Produced by Gerard Arthus, Meredith Bach and the Online
Distributed Proofreading Team at https://www.pgdp.net (This
file was produced from images generously made available
by The Internet Archive)







  +-----------------------------------------------------------------+
  | Transcriber's Note:                                             |
  |                                                                 |
  | All Greek words have been transliterated into English, and are  |
  | contained within { } brackets.                                  |
  +-----------------------------------------------------------------+




             THE CULT OF INCOMPETENCE




          FIRST EDITION       November, 1911.
          SECOND EDITION      July, 1912.




             THE CULT OF INCOMPETENCE


                  By EMILE FAGUET

              _Of the French Academy_

           TRANSLATED FROM THE FRENCH BY
                  BEATRICE BARSTOW

              WITH AN INTRODUCTION BY
                   THOMAS MACKAY

                     NEW YORK:
              E. P. DUTTON & COMPANY
                       1912




                     CONTENTS.



                                                   PAGE

  INTRODUCTION                                        1

  CHAPTER

    I. THE PRINCIPLES OF FORMS OF GOVERNMENT         12

   II. CONFUSION OF FUNCTIONS                        37

  III. THE REFUGES OF EFFICIENCY                     59

   IV. THE COMPETENT LEGISLATOR                      66

    V. LAWS UNDER DEMOCRACY                          82

   VI. THE INCOMPETENCE OF GOVERNMENT                92

  VII. JUDICIAL INCOMPETENCE                         96

 VIII. EXAMPLES OF INCOMPETENCE                     123

   IX. MANNERS                                      156

    X. PROFESSIONAL CUSTOMS                         162

   XI. ATTEMPTED REMEDIES                           172

  XII. THE DREAM                                    216

       INDEX                                        237




THE CULT OF INCOMPETENCE.

INTRODUCTION.


Though it may not have been possible in the following pages to reproduce
the elegant and incisive style of a master of French prose, not even the
inadequacies of a translation can obscure the force of his argument. The
only introduction, therefore, that seems possible must take the form of
a request to the reader to study M. Faguet's criticism of modern
democracy with the daily paper in his hand. He will then see, taking
chapter by chapter, how in some aspects the phenomena of English
democracy are identical with those described in the text, and how in
others our English worship of incompetence, moral and technical, differs
considerably from that which prevails in France. It might have been
possible, as a part of the scheme of this volume, to note on each page,
by way of illustration, instances from contemporary English practice,
but an adequate execution of this plan would have overloaded the text,
or even required an additional volume. Such a volume, impartially worked
out with instances drawn from the programme of all political parties,
would be an interesting commentary on current political controversy, and
it is to be hoped that M. Faguet's suggestive pages will inspire some
competent hand to undertake the task.

If M. Faguet had chosen to refer to England, he might, perhaps, have
cited the constitution of this country, as it existed some seventy years
ago, as an example of a "demophil aristocracy," raised to power by an
"aristocracy-respecting democracy." It is not perhaps wise in political
controversy to compromise our liberty of action in respect of the
problems of the present time, by too deferential a reference to a golden
age which probably, like Lycurgus in the text, p. 73, never existed at
all, but it has been often stated, and undoubtedly with a certain amount
of truth, that the years between 1832 and 1866 were the only period in
English history during which philosophical principles were allowed an
important, we cannot say a paramount, authority over English
legislation. The characteristic features of the period were a
determination to abolish the privileges of the few, which, however,
involved no desire to embark on the impossible and inequitable task of
creating privileges for the many; a deliberate attempt to extirpate the
servile dependence of the old poor law, and a definite abandonment of
the plan of distributing economic advantages by eleemosynary state
action. This policy was based on the conviction that personal liberty
and freedom of private enterprise were the adequate, constructive
influences of a progressive civilisation. Too much importance has
perhaps been attached to the relatively unimportant question of the
freedom of international trade, for this was only part of a general
policy of emancipation which had a much more far-reaching scope. Rightly
understood the political philosophy of that time, put forward by the
competent statesmen who were then trusted by the democracy, proclaimed
the principle of liberty and freedom of exchange as the true solvents of
the economic problems of the day. This policy remained in force during
the ministry of Sir R. Peel and lasted right down to the time of the
great budgets of Mr. Gladstone.

If we might venture, therefore, to add another to the definitions of
Montesquieu, we might say that the principle animating a liberal
constitutional government was liberty, and that this involved a definite
plan for enlarging the sphere of liberty as the organising principle of
civil society. To what then are we to impute the decadence from this
type into which parliamentary government seems now to have fallen? Can
we attribute this to neglect or to exaggeration of its animating
principle, as suggested in the formula of Montesquieu? It is a question
which the reader may find leisure to investigate; we confine ourselves
to marking what seem to be some of the stages of decay.

When the forces of destructive radicalism had done their legitimate
work, it seemed a time for rest and patience, for administration rather
than for fresh legislation and for a pause during which the principles
of liberty and free exchange might have been left to organise the
equitable distribution of the inevitably increasing wealth of the
country. The patience and the conviction which were needed to allow of
such a development, rightly or wrongly, were not forthcoming, and
politicians and parties have not been wanting to give effect to
remedies hastily suggested to and adopted by the people. Political
leaders soon came to realise that recent enfranchisements had added a
new electorate for whom philosophical principles had no charm. At a
later date also, Mr. Gladstone, yielding to a powerful and not
over-scrupulous political agitation, suddenly determined to attempt a
great constitutional change in the relations between the United Kingdom
and Ireland. Whether the transference of the misgovernment of Ireland
from London to Dublin would have had results as disastrous or as
beneficial as disputants have asserted, may be matter for doubt, but the
manner in which the proposal was made certainly had one unfortunate
consequence. Mr. Gladstone's action struck a blow at the independence
and self-respect, or as M. Faguet terms it, the moral competence of our
parliamentary representation from which it has never recovered. Men were
called on to abandon, in the course of a few hours, opinions which they
had professed for a lifetime and this not as the result of conviction
but on the pressure of party discipline. Political feeling ran high.
The "Caucus" was called into more active operation. Political parties
began to invent programmes to capture the groundlings. The conservative
party, relinquishing its useful function of critic, revived the old
policy of eleemosynary doles, and, in an unlucky moment for its future,
has encumbered itself with an advocacy of the policy of protection. For
strangely enough the democracy, the bestower of power, though developing
symptoms of fiscal tyranny and a hatred of liberty in other directions
clings tenaciously to freedom of international trade--for the present at
least--and it would seem that the electioneering caucus has, in this
instance, failed to understand its own business. The doles of the new
State-charity were to be given to meet contributions from the
beneficiaries, but as the class which for one reason or another is ever
in a destitute condition, could not or would not contribute, the only
way in which the benevolent purpose of the agitation could be carried
out was by bestowing the dole gratuitously. The flood gates, therefore,
had to be opened wider, and we have been and still are exposed to a rush
of philanthropic legislation which is gradually transferring all the
responsibilities of life from the individual to the state. Free trade
for the moment remains, and it is supposed to be strongly entrenched in
the convictions of the liberal party. Its position, however, is
obviously very precarious in view of the demands made by the militant
trade unions. These, in their various spheres, claim a monopoly of
employment for their members, to the exclusion of those who do not
belong to their associations. Logic has something, perhaps not much, to
do with political action, and it is almost inconceivable that a party
can go on for long holding these two contradictory opinions. Which of
them will be abandoned, the future only can tell.

The result of all this is a growing disinclination on the part of the
people to limit their responsibilities to their means of discharging
them, the creation of a proletariate which in search of maintenance
drifts along the line of least resistance, dependence on the government
dole. In the end too it must bring about the impoverishment of the
state, which is ever being called on to undertake new burdens; for the
individual, thus released from obligation to discharge, is still left
free to create responsibilities, for which it is now the business of the
State to make provision. Under such a system the ability to pay as well
as the number of the solvent citizens must continuously decline.

The proper reply to this legislation which we describe as predatory in
the sense that we describe the benevolent habits of Robin Hood as
predatory, cannot be made by the official opposition which was itself
the first to step on the down grade, and which only waits the chances of
party warfare to take its turn in providing _panem et circenses_ at the
charge of the public exchequer. In this way, progress is brought to a
standstill by the chronic unwillingness of the rate- and tax-payers
to find the money. A truer policy, based on the voluntary action of
citizens and capable of indefinite and continuous expansion, finds no
support among politicians, for all political parties seem to be held in
the grip of the moral and technical incompetence which M. Faguet has so
wittily described. The only reply to a government bent on such courses
is that which above has been imputed--perhaps without sufficient
justification--to the governments of the period 1832-1866; and that
reply democracy, as at present advised, will allow no political party to
make.

There does not appear, therefore, to be much difference between the
situation here and in France, and it is very interesting to notice how
in various details there is a very close parallelism between events in
this country and those which M. Faguet has described. The position of
our Lord Chancellor, who has been bitterly attacked by his own party, in
respect of his appointment of magistrates, is very similar to that of M.
Barthou, quoted on p. 118. Our judicial system has hitherto been
considered free from political partisanship, but very recently and for
the first time a minister in his place in parliament, has rightly or
wrongly seen fit to call in question the impartiality of our judicial
bench, and the suspicion, if, as appears to be the case, it is widely
entertained by persons heated in political strife, will probably lead to
appointments calculated to ensure reprisals. Astute politicians do not
commit themselves to an attack on a venerated institution, till they
think they know that that institution is becoming unpopular with the
followers who direct their policy. Criminal verdicts also, especially
on the eve of an election, are now made liable to revision by ministers
scouring the gaols of the country in search of picturesque malefactors
whom, with an accompaniment of much philanthropic speech, they proceed
to set at liberty. Even the first principles of equity, as ordinarily
understood, seem to have lost their authority, when weighed in the
balance against the vote of the majority. Very recently the members of
an honourable and useful profession represented to a minister that his
extension of a scheme of more or less gratuitous relief to a class which
hitherto had been able and willing to pay its way, was likely to deprive
them of their livelihood. His reply, _inter alia_, contained the
argument that the class in question was very numerous and had many
votes, and that he doubted whether any one would venture to propose its
exclusion except perhaps a member for a university; as a matter of fact
some such proposal had been made by one of the university members whose
constituents were affected by the proposal. The minister further
declared that he did not think that such an amendment could obtain a
seconder. The argument seems to impute to our national representatives a
cynical disregard of equity, and a blind worship of numbers, which if
true, is an instance of moral incompetence quite as remarkable as
anything contained in M. Faguet's narrative.

If readers of this volume will take the trouble to annotate their copies
with a record of the relevant incidents which meet them every day of
their lives, they cannot fail to acknowledge how terribly inevitable is
the rise of incompetence to political power. The tragedy is all the more
dreadful, when we recognise, as we all must, the high character and
ability of the statesmen and politicians who lie under the thrall of
this compelling necessity.

This systematic corruption of the best threatens to assume the
proportions of a national disaster. It is the system, not the actors in
it, which M. Faguet analyses and invites us to deplore.

                                                             T. MACKAY.




CHAPTER I.

THE PRINCIPLES OF FORMS OF GOVERNMENT.


The question has often been asked, what is the animating principle of
different forms of government, for each, it is assumed, has its own
principle. In other words, what is the general idea which inspires each
political system?

Montesquieu, for instance, proved that the _principle_ of monarchy is
_honour_, the principle of despotism _fear_, the principle of a republic
_virtue_ or patriotism, and he added with much justice that governments
decline and fall as often by carrying their principle to excess, as by
neglecting it altogether.

And this, though a paradox, is true. At first sight it may not be
obvious how a despotism can fall by inspiring too much fear, or a
constitutional monarchy by developing too highly the sentiment of
honour, or a republic by having too much virtue. It is nevertheless
true.

To make too common a use of fear is to destroy its efficacy. As Edgar
Quinet happily puts it: "If we want to make use of fear we must be
certain that we can use it always." We cannot have too much honour, but
when we can appeal to this sentiment only and when distinctions,
decorations, orders, ribbons--in a word _honours_--are multiplied,
inasmuch as we cannot increase such things indefinitely, those who have
none become as discontented as those who, having some, want more.

Finally we cannot, of course, have too much virtue, and naturally here
governments will fall not by exaggerating but by abandoning their
guiding principle. Yet is it not sometimes true that by demanding from
citizens too great a devotion to their country, we end by exhausting
human powers of endurance and sacrifice? This is what happened in the
case of Napoleon, who, perhaps unwittingly, required too much from
France, for the building up of a 'Greater France.'

But that, some one will object, was not a republic!

From the point of view of the sacrifices required from the citizen, it
was a republic, similar to the Roman Republic and to the French Republic
of 1792. All the talk was 'for the glory of our country,' 'heroism,
heroism, nothing but heroism'! If too much is required of it, civic
virtue can be exhausted.

It is, then, very true that governments perish just as much from an
excess as from a neglect of their appropriate principle. Montesquieu
without doubt borrowed his general idea from Aristotle, who remarks not
without humour, "Those, who think that they have discovered the basis of
good government, are apt to push the consequences of their new found
principle too far. They do not remember that disproportion in such
matters is fatal. They forget that a nose which varies slightly from the
ideal line of beauty appropriate for noses, tending slightly towards
becoming a hook or a snub, may still be of fair shape and not
disagreeable to the eye, but if the excess be very great, all symmetry
is lost, and the nose at last ceases to be a nose at all." This law of
proportion holds good with regard to every form of government.

               *       *       *       *       *

Starting from these general ideas, I have often wondered what principle
democrats have adopted for the form of government which they favour,
and it has not required a great effort on my part to arrive at the
conclusion that the principle in question is the worship and
cultivation, or, briefly 'the cult' of incompetence or inefficiency.

Let us examine any well-managed and successful business firm or factory.
Every employee does the work he knows and does best, the skilled
workman, the accountant, the manager and the secretary, each in his
place. No one would dream of making the accountant change places with a
commercial traveller or a mechanic.

Look too at the animal world. The higher we go in the scale of organic
existence, the greater the division of labour, the more marked the
specialisation of physiological function. One organ thinks, another
acts, one digests, another breathes. Now is there such a thing as an
animal with only one organ, or rather is there any animal, consisting of
only one organ, which breathes and thinks and digests all at the same
time? Yes, there is. It is called the amoeba, and the amoeba is the
very lowest thing in the animal world, very inferior even to a
vegetable.

In the same way, without doubt, in a well constituted society, each
organ has its definite function, that is to say, administration is
carried on by those who have learnt how to administer, legislation and
the amendment of laws by those who have learnt how to legislate, justice
by those who have studied jurisprudence, and the functions of a country
postman are not given to a paralytic. Society should model itself on
nature, whose plan is specialisation. "For," as Aristotle says, "she is
not niggardly, like the Delphian smiths whose knives have to serve for
many purposes, she makes each thing for a single purpose, and the best
instrument is that which serves one and not many uses." Elsewhere he
says, "At Carthage it is thought an honour to hold many offices, but a
man only does one thing well. The legislator should see to this, and
prevent the same man from being set to make shoes and play the flute." A
well-constituted society, we may sum up, is one where every function is
not confided to every one, where the crowd itself, the whole body
social, is not told: "It is your business to govern, to administer, to
make the laws, &c." A society, where things are so arranged, is an
amoebic society.

That society, therefore, stands highest in the scale, where the division
of labour is greatest, where specialisation is most definite, and where
the distribution of functions according to efficiency is most thoroughly
carried out.

                *       *       *       *       *

Now democracies, far from sharing this view, are inclined to take the
opposite view. At Athens there was a great tribunal composed of men
learned in, and competent to interpret, the law. The people could not
tolerate such an institution, so laboured to destroy it and to usurp its
functions. The crowd reasoned thus. "We can interpret and carry out
laws, because we make them." The conclusion was right, but the minor
premise was disputable. The retort can be made: "True, you can interpret
and carry out laws because you make them, but perhaps you have no
business to be making laws." Be that as it may, the Athenian people not
only interpreted and applied its own laws, but it insisted on being paid
for so doing. The result was that the poorest citizens sat judging all
day long, as all others were unwilling to sacrifice their whole time for
a payment of six drachmas. This plebeian tribunal continued for many
years. Its most celebrated feat was the judgment which condemned
Socrates to death. This was perhaps matter for regret, but the great
principle, the sovereignty of incompetence, was vindicated.

Modern democracies seem to have adopted the same principle, in form they
are essentially amoebic. A democracy, well-known to us all, has been
evolved in the following manner.

It began with this idea; king and people, democratic royalty, royal
democracy. The people makes, the king carries out, the law; the people
legislates, the king governs, retaining, however, a certain control over
the law, for he can suspend the carrying out of a new law when he
considers that it tends to obstruct the function of government. Here
then was a sort of specialisation of functions. The same person, or
collective body of persons, did not both legislate and govern.

This did not last long. The king was suppressed. Democracy remained, but
a certain amount of respect for efficiency remained too. The people, the
masses, did not, every single man of them, claim the right to govern and
to legislate directly.

It did not even claim the right to nominate the legislature directly. It
adopted indirect election, _à deux degrés_, that is, it nominated
electors who in turn nominated the legislature. It thus left two
aristocracies above itself, the first electors and the elected
legislature. This was still far removed from democracy on the Athenian
model which did everything itself.

This does not mean that much attention was paid to efficiency. The
electors were not chosen because they were particularly fitted to elect
a legislature, nor was the legislature itself elected with any reference
to its legislative capacity. Still there was a certain pretence of a
desire for efficiency, a double pseudo-efficiency. The crowd, or rather
the constitution, assumed that legislators elected by the delegates of
the crowd were more competent to make laws than the crowd itself.

This somewhat curious form of efficiency I have called _compétence par
collation_, efficiency or competence conferred by this form of
selection. There is absolutely nothing to show that so-and-so has the
slightest legislative or juridical faculty, so I confer on him a
certificate of efficiency by the confidence I repose in him when
nominating him for the office, or rather I show my confidence in the
electors and they confer a certificate of efficiency on those whom they
nominate for the legislature.

This, of course, is devoid of all common sense, but appearances, and
even something more, are in its favour.

It is not common sense for it involves something being made out of
nothing, inefficiency producing efficiency and zero extracting 'one' out
of itself. This form of selection, though it does not appeal to me under
any circumstances, is legitimate enough when it is exercised by a
competent body. A university can confer a degree upon a distinguished
man because it can judge whether his degreeless condition is due to
accident or not. It would, however, be highly ridiculous and paradoxical
if the general public were to confer mathematical degrees. A degree of
efficiency conferred by an inefficient body is contrary to common sense.

There is, however, some plausibility and indeed a little more than
plausibility in favour of this plan. Degrees in literature and in
dramatic art are conferred, given by 'collation,' by incompetent people,
that is by the public. We can say to the public: "You know nothing of
literary and dramatic art." It will retort: "True, I know nothing, but
certain things move me and I confer the degree on those who evoke my
emotions." In this it is not altogether wrong. In the same way the
degree of doctor of political science is conferred by the people on
those who stir its emotions and who express most forcibly its own
passions. These doctors of political science are the empassioned
representatives of its own passions.

--In other words, the worst legislators!--

Yes, very nearly so, but not quite. It is very useful that we should
have an exponent of popular passion at the crest of the social wave, to
tell us not indeed what the crowd is thinking, for the crowd never
thinks, but what the crowd is feeling, in order that we may not cross it
too violently or obey it too obsequiously. An engineer would call it the
science of the strength of materials.

A medium assures me that he had a conversation with Louis XIV, who said
to him: "Universal suffrage is an excellent thing in a monarchy. It is a
source of information. When it recommends a certain course of action it
shows us that this is a thing which we must not do. If I could have
consulted it over the Revocation of the Edict of Nantes, it would have
given me a clear mandate for that Revocation and I should have known
what to do, and that Edict would not have been revoked. I acted as I
did, because I was advised by ministers whom I considered experienced
statesmen. Had I been aware of the state of public opinion I should have
known that France was tired of wars and new palaces and extravagance.
But this was not an expression of passion and prejudice, but a cry of
suffering. As far as passion and prejudice are concerned we must go
right in the teeth of public opinion, and universal suffrage will tell
you what that is. On the other hand we must pay heed, serious heed to
every cry of pain, and here too universal suffrage will come to our aid.
Universal suffrage is necessary to a monarchy as a source of
information."

This, I am told, is Louis XIV's present opinion on the subject.

As far as legislation therefore is concerned, the attempt to secure
competence by 'collation' is an absurdity. Yet it is an inverted sort
of competence useful for indicating the state of a nation's temper. From
this it follows that this system is as mischievous in a republic as it
would be wholesome in a monarchy. It is not therefore altogether bad.

The democracy which we have in view, after having been governed by the
representatives of its representatives for ten years, submitted for the
next fifteen years to the rule of one representative and took no
particular advantage therefrom.

Then for thirty years it adopted a scheme which aimed at a certain
measure of efficiency. It assumed that the electors of the legislature
ought not to be nominated, but marked out by their social position, that
is their fortune. Those who possessed so many drachmas were to be
electors.

What sort of a basis for efficiency is this? It is a basis but certainly
a somewhat narrow one.

It is a basis, first, because a man who owns a certain fortune has a
greater interest than others in a sound management of public business,
and self-interest opens and quickens the eye; and again a man who has
money and does not lose it cannot be altogether a fool.

On the other hand it is a narrow basis, because the possession of money
is of itself no guarantee of political ability, and the system leads to
the very questionable proposition that every rich man is a competent
social reformer. It is, however, a sort of competence, but a competence
very precariously established and on a very narrow basis.

This system disappeared and our democracy, after a short interregnum,
repeated its previous experiment and submitted for eighteen years to the
rule of one delegate with no great cause to congratulate itself on the
result.

It then adopted democracy in a form almost pure and simple. I say
almost, for the democratic system pure and simple involves the direct
government of the people without any intervening representatives, by
means of a continuous plebiscite. Our democracy then set up and still
maintains a democratic system almost pure and simple, that is to say, it
established government of the nation by delegates whom it itself elected
and by these delegates strictly and exclusively. This time we have
reached an apotheosis of incompetence that is well nigh absolute.

This, our present system, purports to be the rule of efficiency chosen
by the arbitrary form of selection which has been described. Just as the
bishop in the story, addressing a haunch of venison, exclaimed: "I
baptise thee carp," so the people says to its representatives: "I
baptise you masters of law, I baptise you statesmen, I baptise you
social reformers." We shall see later on that this baptism goes very
much further than this.

If the people were capable of judging of the legal and psychological
knowledge possessed by those who present themselves for election, this
form of selection need not be prohibitive of efficiency and might even
be satisfactory; but in the first place, the electors are not capable of
judging, and secondly, even if they were, nothing would be gained.

Nothing would be gained, because the people never places itself at this
point of view. Emphatically never! It looks at the qualifications of the
candidate not from a scientific but from a moral point of view.

--Well that surely is something, and, in a way, a guarantee of
efficiency. The legislators are not capable of making laws, it is true;
but at least they are honest men. This guarantee of moral efficiency,
some critic will say, gives me much satisfaction.

Please be careful, I reply, we should never think of giving the
management of a railway station to the most honest man, but to an honest
man who, besides, understood thoroughly railway administration. So we
must put into our laws not only honest intentions, but just principles
of law, politics, and society.

Secondly, if the candidates are considered from the point of view of
their moral worth it is in a peculiar fashion. High morality is imputed
to those who share the dominant passions of the people and who express
themselves thereon more violently than others. Ah! these are our honest
men, it cries, and I do not say that the men of its choice are
dishonest, I only say that by this criterion they are not infallibly
marked out even as honest.

--Still, some one replies, they are probably disinterested, for they
follow popular prejudices, and not their own particular, individual
wishes.

Yes, that is just what the masses believe, while they forget that there
is nothing easier than to simulate popular passion in order to win
popular confidence and become a political personage. If
disinterestedness is really so essential to the people, only those
should be elected who oppose the popular will and who show thereby that
they do not want to be elected. Or better still only those who do not
stand for election should be elected, since not to stand is the
undeniable sign of disinterestedness. But this is never done. That which
should always be done is never done.

--But, some one will say, your public bodies which recruit their numbers
by co-optation, Academies and learned societies, do not elect their
members in this way.--

Quite so, and they are right. Such bodies do not want their members to
be disinterested but scientific. They have no reason to prefer an
unwilling member to one who is eager to be elected. Their point of view
is entirely different. The people, which pretends to set store by high
moral character, should exclude from power those who are ambitious of
power, or at least those who covet it with a keenness that suggests
other than disinterested motives.

These considerations show us what the crowd understands by the moral
worth of a man. The moral worth of a man consists, as far as the crowd
is concerned, in his entertaining or pretending to entertain the same
sentiments as itself, and it is just for this reason that the
representatives of the multitude are excellent as documents for
information, but detestable, or at least, useless, and therefore
detestable, as legislators.

Montesquieu, who is seldom wrong, errs in my opinion when he says, "The
people is well-fitted to choose its own magistrates." He, it is true,
did not live under a democracy. For consider, how could the people be
fitted to choose its own magistrates and legislators, when Montesquieu
himself, this time with ample justification, lays down as one of his
principles that morals should correct climate, and that law should
correct morals, and the people, as we know, only thinks of choosing as
its delegates men who share, in every particular, its own manner of
thinking? Climate can be partially resisted by the people; but if the
law should correct morals, legislators should be chosen who have taken
up an attitude of reaction against current morality. It would be very
curious if such a choice were ever made, and not only is it never made
but the contrary invariably happens.

To sum it all up, it is intellectual incompetence, nay moral
incompetence which is sought instinctively in the people's choice.

              *       *       *       *       *

If possible, it is more than this. The people favours incompetence, not
only because it is no judge of intellectual competence and because it
looks on moral competence from a wrong point of view, but because it
desires before everything, as indeed is very natural, that its
representatives should resemble itself. This it does for two reasons.

First, as a matter of sentiment, the people desires, as we have seen,
that its representatives should share its feelings and prejudices. These
representatives can share its prejudices and yet not absolutely resemble
it in morals, habits, manners and appearance; but naturally the people
never feels so certain that a man shares its prejudices and is not
merely pretending to do so, as when the man resembles it feature by
feature. It is a sign and a guarantee. The people is instinctively
impelled therefore to elect men of the same habits, manners and even
education as itself, or shall we say of an education slightly superior,
the education of a man who can talk, but only superior in a very slight
degree.

In addition to this sentimental reason, there is another, which is
extremely important, for it goes to the very root of the democratic
idea. What is the people's one desire, when once it has been stung by
the democratic tarantula? It is that all men should be equal, and in
consequence that all inequalities natural as well as artificial should
disappear. It will not have artificial inequalities, nobility of birth,
royal favours, inherited wealth, and so it is ready to abolish nobility,
royalty, and inheritance. Nor does it like natural inequalities, that is
to say a man more intelligent, more active, more courageous, more
skilful than his neighbours. It cannot destroy these inequalities, for
they are natural, but it can neutralise them, strike them with impotence
by excluding them from the employments under its control. Democracy is
thus led quite naturally, irresistibly one may say, to exclude the
competent precisely because they are competent, or if the phrase
pleases better and as the popular advocate would put it, not because
they are competent but because they are unequal, or, as he would
probably go on to say, if he wished to excuse such action, not because
they are unequal, but because being unequal they are suspected of being
opponents of equality. So it all comes to the same thing. This it is
that made Aristotle say that where merit is despised, there is
democracy. He does not say so in so many words, but he wrote: "Where
merit is not esteemed before everything else, it is not possible to have
a firmly established aristocracy," and that amounts to saying that where
merit is not esteemed, we enter at once on a democratic regime and never
escape from it.

The chance, then, of efficiency coming to the front in this state of
affairs is indeed deplorable.

First and last, democracy--and it is natural enough--_wishes to do
everything itself_, it is the enemy of all specialisation of functions,
particularly it wishes to govern, without delegates or intermediaries.
Its ideal is direct government as it existed at Athens, its ideal is
"democracy," in the terminology of Rousseau, who applied the word to
direct government and to direct government only.

Forced by historical events and perhaps by necessity to govern by
delegates, how could democracy still contrive to govern directly or
nearly so, although continuing to govern through delegates?

Its first alternative is, perhaps, to impose on its delegates an
imperative mandate. Delegates under this condition become mere agents of
the people. They attend the legislative assembly to register the will of
the people just as they receive it, and the people in reality governs
directly. This is what is meant by the imperative mandate.

Democracy has often considered it, but never with persistence. Herein it
shows good sense. It has a shrewd suspicion that the imperative mandate
is never more than a snare and a delusion. Representatives of the people
meet and discuss, the interests of party become defined. Henceforward
they are the prey of the goddess Opportunity, the Greek {Kairos}.
Then it happens one day that to vote according to their mandate would be
very unfavourable to the interest of their party. They are therefore
obliged to be faithless to their party by reason of their fidelity to
their mandate, or disobedient to their mandate by reason of their
obedience to their party; and in any case to have betrayed their mandate
with this very praiseworthy and excellent intention is a thing for which
they can take credit or at least obtain excuse with the electors--and on
such a matter it will be very difficult to refute them.

The imperative mandate is therefore a very clumsy instrument for work of
a very delicate character. The democracy, instinctively, knows this very
well, and sets no great store by the imperative mandate.

What other alternative is there for it? Something very much finer, the
substance instead of the shadow. It can elect men who resemble it
closely, who follow its sentiments closely, who are in fact so nearly
identical with itself that they may be trusted to do surely,
instinctively, almost mechanically that which it would itself do, if it
were itself an immense legislative assembly. They would vote, without
doubt, according to circumstances, but also as their electors would vote
if they were governing directly. In this way democracy preserves its
legislative power. It makes the law, and this is the only way it can
make it.

Democracy, therefore, has the greatest inducement to elect
representatives who are representative, who, in the first place,
resemble it as closely as possible, who, in the second place, have no
individuality of their own, who finally, having no fortune of their own,
have no sort of independence.

We deplore that democracy surrenders itself to politicians, but from its
own point of view, a point of view which it cannot avoid taking up, it
is absolutely right. What is a politician? He is a man who, in respect
of his personal opinions, is a nullity, in respect of education, a
mediocrity, he shares the general sentiments and passions of the crowd,
his sole occupation is politics, and if that career were closed to him,
he would die of starvation.

He is precisely the thing of which the democracy has need.

He will never be led away by his education to develop ideas of his own;
and having no ideas of his own, he will not allow them to enter into
conflict with his prejudices. His prejudices will be, at first by a
feeble sort of conviction, afterwards by reason of his own interest,
identical with those of the crowd; and lastly, his poverty and the
impossibility of his getting a living outside of politics make it
certain that he will never break out of the narrow circle where his
political employers have confined him; his imperative mandate is the
material necessity which obliges him to obey; his imperative mandate is
his inability to quarrel with his bread and butter.

Democracy obviously has need of politicians, has need of nothing else
but politicians, and has need indeed that there shall be in politics
nothing else but politicians.

Its enemy, or rather the man whom democracy dreads because he means to
govern and does not intend to allow the mob to govern through him, is
the man who succeeds in getting elected for some constituency or other,
either by the influence of his wealth or by the prestige of his talent
and notoriety. Such a man is not dependent on democracy. If a
legislative assembly were entirely or by a majority composed of rich
men, men of superior intelligence, men who had an interest in attending
to the trades or professions in which they had succeeded rather than in
playing at politics, they would vote according to their own ideas, and
then--what would happen? Why then democracy would be simply suppressed.
It would no longer legislate and govern; there would be, to speak
exactly, an aristocracy, not very permanently established perhaps, but
still an aristocracy which would eliminate the influence of the people
from public affairs.

Clearly it is almost impossible for the democracy, if it means to
survive, to encourage efficiency, nay it is almost impossible for it to
refrain from attempting to destroy efficiency.

Thus, we may sum up, only those are elected as the representatives of
the people, who are its exact counterparts and constant dependents.




CHAPTER II.

CONFUSION OF FUNCTIONS.


And what is the result of all this? The result, which is very logical,
very just from the democratic point of view, and precisely that which
the democracy desires and cannot do otherwise than desire, is that the
national representatives do exactly what the people would wish them to
do, and what the people would do itself if it undertook to govern
directly itself. _The representative government wishes to do everything
itself_, just as the people would like to do, if it were itself
exercising the functions of government directly, just as it did in olden
times on the Pnyx at Athens.

Montesquieu realised this fully, though naturally he had no experience
of how the theory worked under a representative and parliamentary
system. The principle of it all is at bottom the same, and only the
change of a single phrase is needed to make the following quotation
strictly applicable. "The principle of democracy," he says, "is
perverted not only when it loses the spirit of equality, but still more
_when it carries the spirit of equality to an extreme, and when every
one wishes to be the equal of those whom he chooses to govern him_. For
then the people, not being able to tolerate the authority which it has
created, _wishes to do everything itself_, to deliberate for the Senate,
to act for the magistrates, and to usurp the functions of the judges.
The people wishes to exclude the magistrates from their functions, and
the magistrates naturally are no longer respected. The deliberations of
the Senate are allowed to have no weight, and senators naturally fall
into contempt."

Let us translate the foregoing passage into the language of to-day.
Under democratic parliamentary government the representatives of the
people are determined to do everything themselves. They must be equal to
those whom they choose for their rulers. They cannot tolerate the
authority which they have entrusted to the Government. They must
themselves govern in the place of the Government, administer in the
place of the executive staff, substitute their own authority for that
of all the bench of judges, perform the duties of magistrates, and, in a
word, throw off all regard and respect for persons and things.

This is the true inwardness of the popular spirit, the will of the
people which wishes to do everything itself, or what is the same thing,
through its representatives, its faithful and servile creatures.

From this point onwards efficiency is hunted and exterminated in every
direction; just as it was excluded in the election of representatives,
so the representatives laboriously and continuously exclude it from
every sort of office and employment under the public service.

The Government, to begin our analysis of functional confusion at the
top, ought to be watched and advised by the national representatives,
but it ought to be independent of the national representatives, at least
it ought not to be inextricably mixed up with them, in other words the
national representatives ought not to govern. Under democracy this is
precisely what they want to do. They elect the Government, a privilege
which need not be denied them; but, "not being able to tolerate the
authority which they have created," as soon as they have set it up, they
put pressure on it and insist on governing continuously in its place.
The assembly of national representatives is not a body which makes laws,
but a body which, by a never ending string of questions and
interruptions, _dictates_ from day to day to the Government what it
ought to do, that is to say, it is a body which governs.

The country is governed, literally, by the Chamber of Deputies. _This is
absolutely necessary_ if, as the true spirit of the system requires, the
people is to be governed by no one but itself, if there is to be no will
at work other than the will of the people, emanating from itself and
bringing back a sort of harvest of executive acts. Again, I repeat, this
is absolutely necessary, in order that there shall be nothing, not even
originating with the people, which, for a single moment and within the
most narrowly defined limits, shall exercise the functions of
sovereignty over the sovereign people.

This is all very well, but government is an art and we assume that there
is a science of government, and here we have the people governed by
persons who have neither science nor art, and who are chosen precisely
because they have not these qualifications and on the guarantee that
they have none of them!

Again, in a democracy of this kind, if there exist, as a result of
tradition or of some necessity arising out of foreign relations, an
authority, independent for a certain term of years of the legislative
assembly, which has no accounts to render to it and which cannot be
questioned or constitutionally overthrown, that authority is so strange,
and, if the phrase may pass, so monstrous an anomaly, that it dares not
exercise its power, and dreads the scandal which it would raise by
acting on its rights, and seems as it were paralysed with terror at the
very thought of its own existence.

And its attitude is right; for if it exercised its powers, or even lent
itself to any appearance of so doing, there at once would be an act of
will which was not an act of the popular will, a theory altogether
contrary to the spirit of this system. For in this system the chief of
the state can only be the nominal chief of the state. A will of his own
would be an abuse of power, an idea of his own would be an
encroachment, and a word of his own would be an act of high treason.

It follows that, if the constitution has formally conferred these
powers, the constitution on these points is a dead letter, because it
contravenes an unwritten constitution of higher authority, viz., the
inner inspiration of the political institution.

One of these honorary chiefs of the state has said: "During all my term
as president, I was constitutionally silent." This is not correct, for
the constitution gave him leave to speak and even to act. At bottom it
was true, for the constitution, in allowing him to act and speak, was
acting unconstitutionally. In speaking he would have been
constitutional, in holding his tongue he was _institutional_. He had
been in fact _institutionally_ silent. He disobeyed the letter of the
constitution, but he had admirably extracted its meaning from it, and
understood and respected its spirit.

Under democracy, then, the national representatives govern as directly
and as really as possible, dictating a policy to the executive and
neutralising the supreme chief of the executive to whom it is not able
to dictate.

The national representatives are not content with governing, they wish
to administer. Now consider how it would be if the permanent officials
of finance, justice and police, etc., depended solely on their
parliamentary chiefs, who are ministers only because they are the
creatures of the popular assembly, liable to instant and frequent
dismissal; surely then, these officials, more permanent than their
chiefs, would form an aristocracy, and would administer the state
independently of the popular will and according to their own ideas.

This, of course, must not be allowed to happen. There must not be any
will but the people's will, no other power, however limited, but its
own.

This causes a dilemma which is sufficiently remarkable. Here we seem to
have contrary results from the same cause. Since the popular assembly
governs ministers, and frequently dismisses them, they are not able to
govern their subordinates as did Colbert and Louvois, and these
subordinates accordingly are very independent; so it comes about that
the greater the authority which the popular assembly wields over
ministers, the more it is likely to lose in its control over the
subordinates of ministers, and in destroying one rival power it creates
another.

The dilemma, however, is avoided easily enough. No public official is
appointed without receiving its _visa_, and it contrives even to elect
the administrative officials. In the first place, the national
representatives, in their corporate capacity, and in the central offices
of government, watch most attentively the appointment of the permanent
staff, and further each single member of the representative government
in his province, in his department, in his _arrondissement_ picks and
chooses the candidates and really appoints the permanent staff. This is,
of course, necessary, if the national will is to be paramount here as
well as elsewhere, and if the people is to secure servants of its own
type, if it is "to choose its own magistrates," as Montesquieu said.

The people, then, chooses its servants through the intervention of its
representatives; and consider, to return to our point, how absolutely
necessary it is for it to secure representatives who are intellectually
the exact image and imitation of itself. Everything dovetails neatly
together.

Here then we have the people interfering influentially in the
appointment of the civil service. It continues "to do everything
itself." Complaints are raised on all sides of this confusion of
politics with the business of administration, and indeed we hear
continually that politics pervade everything. But what is the reason of
this? It is the principle of the national sovereignty asserting itself.
Politics, political power, means the will of the majority of the nation,
and is it not fitting that the will of the majority should make itself
felt--indeed need we be surprised that it insists on making itself
felt--in the details of public business, as administered by the
permanent staff, as well as elsewhere? The ideal of democracy is that
the people should elect its own rulers, or, if this is not its ideal, it
is its idea, and this is what it does under a parliamentary democracy
through the intervention of its representatives.

This is all very well, but efficiency has been dealt another blow. For
how is a candidate to recommend himself for an office to which
appointment is made by the people and its representatives? By his merit?
His chiefs and his fellow civil servants might be good judges of that;
but the people or its representatives are much less capable of judging.

"The people is admirably fitted to choose those to whom it has to
entrust some part of its authority"; so Montesquieu; we must now examine
this saying a little more closely. What reasons does the philosopher
give? "The people can only be guided by things of which it cannot be
ignorant, and which fall, so to speak, within its own observation. It
knows very well that a man has experience in war, and that he has had
such and such successes; it is therefore quite capable of electing a
general. It knows that a judge is industrious, that many of those who
are litigants in his court go away satisfied, and that he has never been
convicted of bribery, and this is enough to warrant it in appointing to
any judicial office. It has been impressed by the magnificence or riches
of some citizen, and this fits it for appointing an ædile. All these
things are matters of fact about which the man in the street has better
knowledge than the king in his palace."

This passage, I confess, does not appear to be convincing. Why should
not a king in his palace know of the riches of a financier, the
reputation of a judge or the success of a colonel just as well as the
man in the street? There is no difficulty in getting information about
such things. The people knows that such an one was always a good judge
and such another always an excellent officer. Therefore it is qualified
to appoint a general or a high-court judge or other officer of the law.
So be it, but for the selection of a young judge or a young and untried
officer what special source of information has the people? I cannot find
that it has any. In this very argument, Montesquieu limits the
competence of the people to the election of the great chiefs, and of the
most exalted magistrates, and indeed further confines the popular
prerogative in this matter to assigning an office and career to one who
has already given proof of his capacity. But for putting the competent
man for the first time in the place where he is wanted, how has the
people any special instinct or information? Montesquieu shows that the
people can recognise ability when it has been proved, but he says
nothing to show that it recognises readily nascent, unproved talent. The
argument of Montesquieu is not here conclusive.

He has been led astray, it seems to me, by his desire to present his
argument antithetically (using the term in its logical sense). What he
really wished to prove was not so much the truth of the proposition that
he was then advancing, but the falsity of quite another proposition. The
question for him, the question which he had in his mind, was as follows:
Is the people capable of governing the state, of taking measures
beforehand, and of understanding and solving the difficulties of home
and foreign affairs? By no means. Then is it fit to elect its own
magistrates? Well, it might do that. Thus he had been led away by this
antithesis so far as to say: Able to govern?--Certainly not! Able to
elect its own magistrates? Admirably! The explanation of the whole
paragraph which I have just quoted lies in the conclusion, which runs as
follows: "All these things are matters of fact about which the man in
the street has better knowledge than the king in his palace. _But_ can
the people pursue a policy and know how to avail itself of the places,
occasions, and times when action will be profitable? No! certainly
not."

The truth is that the people is a little better fitted to choose a
magistrate than to undertake a policy for the gradual humbling of the
House of Austria. But not very much so, as it is only a little more
difficult to humble the House of Austria, than it is to discover the man
who is able to do it.

The masses are particularly incapable of making initial appointments and
of giving promotion in the early stages of a career to those who deserve
it. Yet in a democracy this is what they are constantly doing.

Again, by what means has the candidate for civil service employment, who
is favoured by the people and its representatives, earned their
approval? By his merit, of which the people and its representatives are
very bad judges? No! By what then? By his conformity to the general
views of the people; that is, by the subserviency of his political
opinions. The political opinions of a candidate for civil service
employment are the only things which mark him out to the popular choice
because they are the only subjects on which the people is a good judge.

Yes, but the subserviency of his political opinions may be combined
with real merit. True, but this is a mere matter of chance. The people
is not, perhaps, in this particular matter consciously hostile to
efficiency, rather it is indifferent, or ignores the qualification
altogether. Indeed, there is no great compliment paid to efficiency in
such transactions.

Here is what inevitably happens. The candidate for a permanent
appointment who is not conscious of possessing any particular merit is
not slow to realise that it is by his political opinions that he will
succeed, and he naturally professes those which are wanted. The
candidate who is conscious of merit, very often knowing very well what
less meritorious competitors are about, and not wishing to be beaten,
also professes the same useful opinions. There we have that "infection
of evil," which M. Renouvier has explained so admirably in his _Science
de la Morale_.

First, then, we see how most of the candidates chosen by the mandatories
of the people are incapable; others who are chosen in spite of their
capacity are men of indifferent character; and character, we must admit,
in all or nearly all public careers is a necessary part of efficiency.

There remains a small number of meritorious persons who have never
identified themselves with current political opinions, and who have
slipped into public employment, thanks to some brief moment of
inattention on the part of the politicians. These intruders sometimes
get on by the mere force of circumstances, but they never reach the
highest posts which are always reserved, as indeed is proper and
fitting, for those in whom the people has put its trust.

This is how the people administers as well as governs through the
intervention of the representative system, dictating to ministers the
policy and the details of government.

--I realise, some one here will object, that administrators are
nominated by the people, but I do not see how the affairs of the country
are actually administered by the people.--

Well, I will tell you. In the first place, by nominating officials it is
already far on the road to controlling them, for it infuses into the
body of the permanent civil service the spirit of the people to the
exclusion of every other source of inspiration, and effectually prevents
the civil service from becoming an aristocracy as otherwise it has
always a tendency to do. Next, the people does not confine itself to
electing its administrators, it watches and spies on them, keeps them in
leading strings, and just as the popular representatives dictate to
ministers the details of government, so also they dictate to
administrators the details of administration.

A _préfet_, a _procureur-général_, an engineer-in-chief under democratic
rule is a much harassed man. He has to play his own hand against his
ministerial chief and the deputies of his district. He ought to obey the
minister, but he has also to obey the deputies of the district which he
administers. In this connection curious points arise and situations not
a little complicated. The _préfet_ owes obedience to the deputies and to
the minister, and the minister obeys the deputies, and it might
therefore have been supposed that there was only one will, the will
which the _préfet_ obeyed. But what the minister has to obey is the
general will of the popular representatives, and it is this will that he
transmits for the allegiance of the _préfet_; but then the _préfet_
finds himself colliding against the individual wills of the deputies of
his district. The result is what we may call conflicts of obedience
which have extraordinary interest for the psychologist, but which are
less agreeable for the _préfet_, the engineer-in-chief, or the
_procureur-général_.

We note then, in the first place, how everything concurs to make the
representative of the popular will as incompetent as he is omnipotent.
Incompetent he undoubtedly is, as we have already seen, to start with,
and _if he were not so already_, he would certainly become so by reason
of the trade or rather of the miscellaneous assortment of trades which
are thrust upon him. The surest way of making a man incompetent is to
make him Jack-of-all-trades, for then he will be master of none. In the
next place, the representative of the popular will and spirit, besides
his trade of legislator, has to cross-examine ministers and to dictate
to them the details of their duty, that is to say, he has to busy
himself in all home and foreign politics. He has also to administer, by
choosing and watching administrators and by controlling and inspiring
their actions. Without saying anything of the small individual services
which it is his interest to render to his constituents and which his
constituents are by no means backward in demanding, he looks on himself
as responsible for the conduct of things in general. He becomes a sort
of universal foreman, not a man, but a man-orchestra, a busybody, so
busy that he can apply himself to nothing. He cannot study, or think, or
investigate, or, to speak accurately, acquire any sense at all.

If he be efficient in some particular subject, when he enters on his
public career, he becomes hopelessly inefficient in all subjects after a
few years of public life, and then, void of all individuality, he
remains nothing but a public man, that is, a man representing the
popular will and never thinking, or able to think, of anything but how
to make that will prevail.

And, to press the point again, this is all that is wanted of him; for
can you conceive a representative of the popular will, who had somehow
preserved a measure of competence in financial or judicial
administration, who would prefer, before other candidates, not a
political partisan but a man of merit, knowledge and aptitude, and who
would even approve in an administrator not acts of political partiality
but acts that are just and in conformity with the interests of the
state? Why! Such a man would be a detestable servant in the eyes of
democracy.

Yes, and I have known such a man. He was not wanting in intelligence or
wit and he was honest. A lawyer, he was naturally interested in
politics. For local reasons he had failed to be elected as deputy or as
senator. Tired of fighting, he obtained a judicial appointment by the
influence of his political friends. He became president of court. A case
was brought before him where the accused, a person not perhaps of
altogether blameless life, was clearly not guilty of any indictable
offence. The accused, however, a former _préfet_, appointed by a
government now become very unpopular, and known as a reactionary and an
aristocrat, was pursued by the animosity of the whole democratic
population of the town and province. The president, in the face of
openly expressed hostility in court, acquitted him. In the evening the
president remarked, not without a touch of humour: "There, that serves
them right for not making me a senator!" In other words: "If they had
accepted me as a politician, they would have made me a fool, or at
least paralysed my efficiency. But they would not have it; so here I am,
a man who knows the law and applies it. So much the worse for them!"

"By making a man a slave Zeus took from him half his soul." So Homer. By
making a man a politician, Demos takes from him his whole soul, and in
omitting to make him a politician, it is foolish enough to leave him his
soul.

This is why Demos hates a permanent civil service. An irremovable
magistrate or functionary is a man whom the constitution sets free from
the grip of the populace. An irremovable official is a man enfranchised,
a free man. Demos does not love free men.

This will explain why in every nation where it is paramount, democracy
suspends from time to time the irremovable independent official element
wherever it is found. The object is nominally to clarify and filter the
_personnel_ of the official world; but really it is intended to teach
the officials whom it spares, that their permanence is only very
relative and that, like every one else, they have to reckon with the
sovereignty of the people which will turn and rend them if they
venture to be too independent.

According to the constitution of 1873 there were irremovable senators in
France. In the interest of good government, this was perhaps a sound
arrangement. The irremovable senators, in the scheme of the
constitution, were intended to be, and in fact were, political and
administrative veterans from whose knowledge, efficiency and experience
their colleagues were to profit. The plan, from this point of view,
might have worked well if the irremovable senators had not been elected
by their colleagues but had become so by right; for example every former
President of the Republic, every former president of the _Cour de
Cassation_, every former president of the Court of Appeal, every
admiral, every archbishop might _ex officio_ have been raised to the
rank of senator for life. From the democratic point of view, however, it
was regarded as a positive outrage that there should exist a
representative of the people who had not to render account to the
people, a representative of the people who had nothing to fear from the
accidents of re-election, no risk of failing to secure re-election, in
other words that a man should be elected for his supposed efficiency,
in no sense representing the people but himself alone.

Permanent senators were abolished. Obviously they constituted a
political aristocracy, founded on the pretence of services rendered, and
the Senate which elected them also fell under the taint of aristocratic
leanings since at that time it recruited its members by co-optation.
This of course could not be tolerated.




CHAPTER III.

THE REFUGES OF EFFICIENCY.


Will efficiency then, you may well ask, when driven out of all public
employment, find refuge somewhere? Certainly it will. In private
employments and in employments paid by public companies. Barristers,
solicitors, doctors, business men, manufacturers and authors are not
paid by the state, nor are engineers, mechanics, railway employees; and
so far from their efficiency being a bar to their employment, it is
their most valuable asset. When a man consults his lawyer or his medical
adviser he obviously has no interest in their politics, and when a
railway company chooses an engineer, it enquires into his qualifications
and ability and is quite indifferent as to whether his political views
coincide with the general mentality of the people.

It is for this reason, or at least partly for this reason, that
democracy tries to nationalise all employment, as a step in the
direction of the nationalisation of everything. For instance it can
partly nationalise the medical profession by establishing appointments
for doctors, at relief offices, schools, and _lycées_. It can also
partly nationalise the legal profession by appointing state-paid
professors of law.

Already the State has considerable control over this class of person,
for most of them have relations in government employment, whom they do
not wish to bring into bad odour by seeming hostile to the opinions of
the majority. The State, however, wants to hold them in still tighter
control by seizing every opportunity of nationalising and socialising
them more completely.

The State wants also to destroy all large associations, and to absorb
their activities. The state purchase of a railway, for instance, is, in
the first place, a means of exploiting the company; for there is always
a hope that the State will be able to filch something out of the
transaction; but its chief recommendation lies in the fact that it
suppresses a whole army of the company's officials and employees, who
were under no obligation to please the Government, and who had no other
interest but to do their work properly. The State will thus transform
this free population into government employees, whose primary duty is to
be docile and subservient.

Under the extreme form and under the complete form of this regime, that
is to say under socialism, everyone will be a government official.

Consequently, say the Socialist theorists, all the alleged drawbacks
above mentioned will disappear. The State, the democracy, the dominant
party, whatever you choose to call it, will no longer be obliged to
select its servants, as you say it does, by reason of their subservience
and their incompetence, because every citizen will be an official. Thus
too will disappear that dual social system, under which half the
population lives on the State, while the other half is independent, and
prides itself on its superiority in character, in intelligence and in
efficiency. Socialism solves the problem.

I do not agree. Under socialism, the electoral system, and, therefore,
the party system will still exist. The citizens will choose the
legislators, the legislators will choose the Government, and the
Government will choose the directors of labour and the distributors of
the means of subsistence. Parties, that is, combinations of interests,
will still exist, and each party will want to capture the legislature in
order to secure the election, from its own number, of the directors of
labour and the distributors of the means of subsistence. These directors
and distributors will be the new aristocrats of socialism, and they will
be expected to arrange "soft jobs" and ampler rations for the members of
their own group or party.

Except that wealth and the last vestiges of liberty have been
suppressed, nothing has been changed, and all the objections above
mentioned still hold. There is no solution here.

If it were a solution, then the socialist government could not long
remain elective. It would have to reign by divine right, like the
Jesuits in Paraguay. It would have to be a despotism, not only in its
policy but in its origin, in fact a monarchy. No intelligent king has
any inducement to choose incompetent men as his officials. His interest
would lead him to do exactly the opposite. You will say that an
intelligent king is a very rare, even an abnormal thing. I readily
agree. Except in a very few instances, which history records with
amazement, a king has exactly the same reasons as the people for
selecting as his favourites men who will not eclipse nor contradict him,
and who consequently seldom turn out to be the best of citizens either
in respect of intelligence or character. Elective socialism and despotic
socialism have the same faults as democracy as we understand the term.

Besides, in truth, the drift of democracy towards socialism is nothing
but a reversion to despotism. If socialism were established, it would
begin by being elective, and as every elective system lives and breathes
and has its being in the party system, the dominant party would elect
the legislature, consequently it would constitute the Government and
would extort from that Government, simply because it has the power to
extort it, every conceivable form of privilege. Exploitation of the
country by the majority would result, as in every country where elective
government prevails.

A socialist government therefore is primarily an oligarchy of directors
of labour and distributors of subsistence. It is a very close oligarchy,
for those beneath it are quite defenceless, levelled down to an
equality of poverty and misery. It is a form of government very
difficult to replace, for it holds in its hands the threads of such an
intricate organisation that it must be protected against crude attempts
to change it, and so it tends to be a permanent oligarchy. It would
therefore concentrate very quickly round a leader, or at any rate,
relegate to the second rank the national representatives and the
electorate.

Such a course of events would be very similar to what occurred under the
First Empire in France, when the military caste eclipsed and domineered
over everything. It became continuously necessary to the State, and
though that necessity passed away, it was soon recalled. The caste then
closed its ranks round the leader who gave it unity, and the strength of
unity.

               *       *       *       *       *

So under socialism, more slowly and perhaps after the lapse of a
generation, the directors of labour and the distributors of food,
peaceful Janissaries of the new order, would form themselves into a
caste, very close, very coherent, and (unlike legislators for whom an
executive council can always be substituted), quite indispensable, and
would close their ranks round a chief who would give them unity and the
strength of unity.

Before we knew socialism, we used to say that democracy tended naturally
to despotism. The situation seems somewhat changed, and we might now say
that it tends to socialism: really nothing has changed. For in tending
towards socialism it is towards despotism that it tends. Socialism is
not conscious of this, for it imagines that it is journeying towards
equality, but out of these utopias of equality it is ever despotism that
emerges.

But this is a digression which refers to the future; let us return to
the matter in hand.




CHAPTER IV.

THE COMPETENT LEGISLATOR.


Democracy, in its modern form, encroaches first upon the executive and
then upon the administrative authorities, and reduces them to subjection
by means of its delegates, the legislators, whom it chooses in its own
image, that is to say, because they are incompetent and governed by
passion, just as in the words of Montesquieu, though he perhaps
contradicts himself a little: "The people is moved only by its
passions."

What ought then the character of the legislator to be? The very
opposite, it seems to me, of the democratic legislator, for he ought to
be well informed and entirely devoid of prejudice.

He ought to be well informed, but his information should not consist
only of book learning, although an extensive legal knowledge is of the
greatest use, as it will prevent him from doing, as so often happens,
the exact opposite of what he intends to do. He should also understand
intimately the temperament and character of the people for whom he
legislates.

For a nation should only be given the laws and commandments that it can
tolerate, as Solon said: "I have given them the best laws that they
could endure," and the God of Israel said to the Jews: "I have given you
precepts which are not good," that is to say, they have only the
goodness which your wickedness will tolerate. "This is the sponge," says
Montesquieu, "which wipes out all the difficulties that can be raised
against the laws of Moses."

The legislator, then, ought to understand the temperament and genius of
the people because he has to frame its laws. As the Germans say, he
ought to be an expert on the psychology of races. Further, he ought to
understand the temperament, peculiarities and character of the people,
without sharing its temperament himself. For where the passions and
inclinations are concerned, experience is not knowledge. On the
contrary, experience prevents us from really knowing; and indeed one of
the conditions of knowledge is absence of an experience which may be
another word for bias.

The ideal legislator, or indeed any legislator worthy of the name, ought
to understand the general tendencies of his people, but he ought to be
able to view them from a position of detachment and to be able to
control them, because it is his business partly to satisfy and partly to
combat these tendencies.

_He has partly to satisfy them_, or at least, to consider them, because
a law which outraged the national temperament would be like Roland's
mare, which had every conceivable good quality with this one serious
defect, that she was dead, and born dead. Suppose the Romans had been
given an international law decreeing respect for conquered peoples, it
would have been a dead letter, and by a sort of contagion it would have
led to the neglect of other laws. Suppose the French were given a
liberal law, a law prescribing respect for the individual rights of the
man and the citizen. Liberty, the object of such a law, is for the
French, as Baron Joannès has remarked: "The right of each man to do what
he likes and to prevent other men from doing what they like." In France
such a law would never obtain any but a very grudging allegiance, and it
would certainly lead to the neglect of other laws.

The legislator ought therefore to understand the natural idiosyncrasies
of his people in order to know how far he dare venture to oppose them.

_Partly he must combat them_, because law should be to a nation, or
otherwise it is merely a police regulation, what the moral law is to an
individual. Law should be a restraint imposed continuously in the hope
of future improvements. It should be a curb on dangerous passions and
injurious desires. It should aid the warfare of enlightened selfishness
against the selfishness of which all are ashamed. That is what
Montesquieu meant when he said that morals should correct climate, and
laws should correct morals.

The law, therefore, to a certain extent should correct national
tendencies, it should be loved a little because it is felt to be just,
feared a little because it is severe, hated a little because it is to a
certain degree out of sympathy with the prevalent temper of the day, and
respected because it is felt to be necessary.

This is the law that the legislator has to frame, and therefore he ought
to have expert knowledge of the genius of the people for whom he
legislates. He must understand both those tendencies which will resist
and those which will welcome him. He must know how far he can go
unopposed and how much he can venture without forfeiting his authority.

This is the principal and essential qualification for the legislator.

The second, as we said before, is that he must be impartial. The very
essence of the legislator is that he should have moderation, that virtue
on which Cicero set so high a value, which is so rare, if we look to its
real meaning, _the perfect balance of soul and mind_. "It seems to me,"
said Montesquieu: "_and I have written this book solely to prove it_,
that the spirit of moderation is essential in a legislator, for
political, as well as moral right, lies between two extremes."

Nothing is more difficult for a man than to control his passions, or
more difficult for a legislator than to control the passions of the
people of whom he forms a part, to say nothing of his own. "Aristotle,"
says Montesquieu, "wanted to gratify, first, his jealousy of Plato and
then his love for Alexander. Plato was horrified at the tyranny of the
Athenians. Machiavel was full of his idol, the Duke of Valentinois.
Thomas More, who was wont to speak of what he had read rather than of
what he had thought, wanted to govern every state upon the model of a
Greek city. Harrington could think of nothing but an English republic,
while hosts of writers thought confusion must reign wherever there was
no monarchy. Laws are always in contact with the passions and prejudices
of the legislator, whether these are his alone, or common to him and to
his people. Sometimes they pass through and merely take colour from the
prejudice of the day, sometimes they succumb to it and make it part of
themselves."

This is just the opposite of what should be. The legislator should be to
the people what conscience is to the heart of the individual. He should
understand its besetting passions in all their bearings and not be
deceived by subterfuge or hypocrisy. Sometimes he must attack them
boldly, sometimes play off one against another, or favour one at the
expense of another which is less influential, now yielding ground, now
recovering it, but he must ever be skilful and impartial and never be
intimidated, diverted from his purpose, nor deceived by his natural
enemies.

He should be, so to speak, more conscientious than conscience itself,
because he must never forget that he has to obey to-morrow the law which
he makes to-day--_semel jussit semper paruit_. He must, therefore, be
absolutely disinterested, a thing most difficult for him, but for which
conscience requires no effort.

Not only must he be without passion, but he must have trained himself to
be impervious to passion, which is much more. We must conceive of him as
a conscience that has risen from the ashes of passion.

As Rousseau said, "to discover the perfect ruler for human society we
must find a superior intelligence who has seen all the passions of man
but has experienced none of them, who has had no sort of relations with
our nature but who knows it to the core, whose happiness is not
dependent on us, but who wishes to promote our welfare, in a word, one
who aims at a distant renown, in a remote future, and who is content to
labour in one age and to enjoy in another."

This is why the ingenious Greeks imagined certain legislators going into
exile to some remote and unknown retreat, as soon as they had made the
people adopt and swear obedience to their laws until their return. It
may have been to bind the citizens by this oath, but is it not equally
probable that they wished to escape from the laws which they themselves
had made? Possibly they felt that they could make them all the stricter
with the prospect of being able to evade obedience of them by flight.

Proudhon said: "I dream of a republic so liberal that in it I shall be
guillotined as a reactionary." Lycurgus was perhaps like Proudhon, in
that he founded so severe a republic that he knew he could not live
under it and resolved to leave it as soon as it was established. Solon
and Sylla remained in the states to which they had given laws; we must
therefore place them higher than Lycurgus who has perhaps this excuse
for himself that in all probability he never existed at all.

But the legend remains to show that the legislator should be so
superior to his own passions and to the passions of his people, that, as
legislator, he should make laws before which, as a man, he should stand
in awe.

This moderation, in the sense in which we use the term, has sometimes
led the legislator to suggest or insinuate laws rather than impose them.
This is not always possible, but it is so occasionally. Montesquieu
tells us the following of St. Louis: "Seeing the manifold abuses of
justice in his time he endeavoured to make them unpopular. He made many
regulations for the courts in his own domain, and in those of his
barons, and he was so successful, that only a short time after his death
his methods were adopted in the courts by many of his nobles. Thus this
prince attained his object, although his regulations were not
promulgated as a general law for the whole kingdom, but merely as an
example which any one might follow in his own interest. He got rid of an
evil by making patent the better way. When men saw in his courts and in
those of his nobles more reasonable and natural forms of procedure, more
conformable to religion and morality, more favourable to public
tranquillity and to the security of persons and property, they adopted
the substance and abandoned the shadow. _To suggest where you cannot
compel, to guide where you cannot demand, that is the supreme form of
skill._"

Montesquieu adds with some optimism though no doubt the idea is
encouraging: "Reason has a natural empire, we resist it, but it triumphs
over our resistance; we persist in error for a time but we always have
to return to it."

The instance above quoted is very remote, and can hardly be applied to
anything in our day. But consider, for instance, the law of Sunday
observance which has been revived from the ecclesiastical law. It was a
mistake to include it in the Code because it was antagonistic to many
French customs, and, in many ways, to the national temperament. The
result is what might have been expected, namely, that it has only been
carried out in rare instances, and with an infinity of trouble. It might
have been made the subject of an edict without being included in the
Code. The State might have given a holiday on Sunday to all its
officials, employees and workmen. It might have been made quite clear
simply by a circular from the Minister of Justice that a workman would
not be punished for breach of contract by refusing to work on a Sunday.
The law of a weekly day of rest would then have existed, without being
formally promulgated, and would have been limited precisely where it
should be, by agreement between masters and men who would submit to
working on Sundays when they saw that it was necessary and inevitable.
Moreover this law would be strong enough to modify without destroying
the ancient customs of the people.

Here is another instance which occurs within the law laid down by the
Code, where the legislator makes use of a method of suggestion and
recommendation. Early in the nineteenth century the legislator
considered that it was seemly for a husband who surprised his wife in
adultery to kill both her and her accomplice. The sentiment is perhaps
questionable, but at all events, it was current. Was it given legal
sanction? No, not precisely. It is inserted in the law in the form of an
insinuation, a discreet recommendation and affectionate encouragement.
The legislator wrote these words: "In _flagrante delicto_ murder is
excusable." I am not approving the sentiment, but only this manner of
indicating rather than enforcing the law and what is thought to be a
wholesome practice, and in other instances I should think it excellent.

Finally, one of the essential qualities of the legislator is to show
discretion in changing existing laws, and for this purpose he should be
immune from the passions of men or at all events complete master of
those which beset him. For law has no real authority unless it is
ancient. Where a law is merely a custom which has become law, it is
invested with considerable authority from the first, because it gains
strength by the antiquity of the original custom. When on the other hand
a law is not an old custom but runs counter to custom, then, before it
can have any authority, it must grow old and become a custom itself.

In both cases it is on its antiquity that the law must depend for its
strength. The law is like a tree, at first it is a tender sapling, then
it grows up, its bark hardens, and its roots go deep into the ground and
cling to the rock.

We ought to consider carefully before we venture to replace the forest
tree by the young sapling. "Most legislators," said Usbek to Rhédi,[A]
"have been men of limited abilities, owing their position to a stroke of
fortune, and consulting nothing but their own whims and prejudices. They
have often abolished established laws quite unnecessarily, and plunged
nations into the chaos that is inseparable from change. It is true that,
owing to some odd chance arising out of the nature rather than out of
the intelligence of mankind, it is sometimes necessary to alter laws,
but the case is very rare and when it does arise it should be handled
with a reverent touch. When it is a question of changing the law, much
ceremony should be observed, and many precautions taken, in order that
the people may be naturally persuaded that laws are sacred things, and
that many formalities must precede any attempt to alter them."

In this passage, as so often elsewhere, Montesquieu is quite
Aristotelian, for Aristotle wrote: "It is evident that at times certain
laws must be changed, but this requires great circumspection for, when
there is little to be gained thereby, inasmuch as it is dangerous that
citizens should be accustomed to find it easy to change the law, it is
better to leave a few errors in our magisterial and legislative
arrangements than to accustom the people to constant change. The
disadvantage of having constant changes in the law is greater than any
risk that we run of contracting a habit of disobedience to the law." For
the law assuredly will be disobeyed, if we regard it as ephemeral,
unstable, and always on the point of being changed.

Some knowledge of the laws of the most important nations, a profound
knowledge of the temperament, character, sentiments, passions, opinions,
prejudices and customs of the nation to which he belongs, moderation of
heart and mind, judgment, impartiality, coolness, nay even a measure of
stolidity, these are the attributes of the ideal legislator. Rather they
are the necessary qualifications of every man who purposes to frame a
good law; they are, indeed, the elementary attributes of a legislator.

We have seen that it is the very opposite quality that democracy likes
and expects of its legislators. It selects incompetent and almost
invariably ignorant men, I have explained why; and its nominees are of a
double distilled incompetence in that their passions would certainly
neutralise their efficiency if they possessed any.

Further we have to observe this curious fact. So entirely does democracy
choose its legislators, because they are dominated by passion, and not
in spite of the fact, chooses them indeed precisely for the reasons for
which it ought to reject them, that any moderate, clear-headed,
practical man who wants to be elected and make use of his powers, has to
start by dissembling his moderation, and by making a noisy display of
factious violence. If he wants to be nominated to a post where it will
be his business to defend and guarantee public security, he has to begin
by advocating civil war: to become a peacemaker he must first pose as a
rebel.

Every popular favourite passes through these two phases, and has to
complete one stage before he starts on the next. Is it not better, you
will ask, that a man's whole career should be spent in defence of law
and order rather than the latter part of it? Not at all, because you
cannot exercise any influence as a friend of law and order unless you
have begun as an anarchist.

These changes of opinion occur so frequently that they merely raise a
smile. They have, however, this drawback, that the friend of law and
order, with a seditious past, never has an undisputed authority, and he
spends half his time explaining the reasons for his defection, and this
is a sore let and hindrance to his subsequent career.

The people always elects men swayed by real or simulated passion. These
will either always remain in a state of frenzied excitement, and they
are the great majority, or they will become moderate men, largely
disqualified and handicapped, as we have above shown, for their new
career. The vast majority of these sentimentalists rush into politics
instead of studying them with deliberation, judgment and wisdom. The
canons of good government as above set out are entirely subverted. The
law does not control and restrain the passions of the populace.
Legislation becomes little more than an expression of their frenzy, a
series of party measures levelled by one faction against the other. The
introduction of a bill is a challenge; the passing of an act is a
victory; definitions which at once damn the legislator, and convict the
system.


[A] Characters in Montesquieu's _Lettres Persanes_. Letter
cxxix.




CHAPTER V.

LAWS UNDER DEMOCRACY.


The truth of my contention is proved by the fact that nowadays all our
laws are emergency laws, a thing that no law should ever be. Montesquieu
advised people to be very chary and to think twice before they destroyed
old laws or pulled down an old house to run up a tent, but his advice is
completely ignored. New laws are made for every change in the weather,
for every little daily incident in politics. We are getting used to this
hand-to-mouth legislation. Like the barbarian warrior, of whom
Demosthenes tells us, who always protected that portion of his person
which had just received a blow, holding his shield up to his shoulder,
when his shoulder had been struck, down again to his thigh when the blow
fell there, the dominant faction only makes laws to protect itself
against an adversary who is, or is thought to be, already in the field,
or it introduces a hurried, ill-digested reform under the pressure of an
alleged scandal.

If an aspirant to the tyranny, as they used to say in Athens, is
nominated deputy in too many constituencies, instantly a law is passed
prohibiting multiple candidatures. For the same reason, for fear of the
same man, _scrutin de liste_ is hurriedly replaced by _scrutin
d'arrondissement_.[B]

If an accused woman is supposed to have been ill-treated at her
examination, taken too abruptly before the interrogatory of the
president, or if the counts are ineptly set out by the public
prosecutor, instantly the whole of the criminal procedure is radically
reformed.

It is the same everywhere. The legislative workshops turn out only "the
latest novelties" of the season. Or perhaps a newspaper would be a still
better simile. First there is the 'interpellation,'[C] once at least
every day; that corresponds to the leading article. Then there are
questions for ministers on this, that and the other trivial occurrence;
that is the serial or short story. Then there is a bill brought in about
something that happened the night before, that is the special article.
Then some deputy assaults his neighbour, this is the general news
column.

You could not have a more faithful representation of the country.
Everything that happens in the morning is dealt with in the evening as
it might be in the village pot-house. The legislative chamber is an
exaggerated reflection of the gossiping public. Now it ought not to be a
copy of the country, it ought to be its soul and brain. But when a
national representative assembly represents only the passions of the
populace it cannot be otherwise than what it is.

In other words modern democracy _is not governed by laws_ but by
decrees, for emergency laws are no better than decrees. A law is an
ancient heritage, consecrated by long usage, which men obey without
stopping to think whether it be law or custom. It forms part of a
coherent, harmonious and logical whole. A law improvised for an
emergency is merely a decree. This is one of the things that Aristotle
saw better than any one. He comments frequently upon the essential and
fundamental distinction between the two, and explains how it is as
dangerous to misunderstand as to ignore it. I quote the passage in which
he brings this out most forcibly: "A fifth form of democracy is that in
which not the law but the multitude has the supreme power, and
supersedes the law by its decrees. This is a state of affairs brought
about by the demagogues. For in democracies which are subject to the
law, the best citizens hold the first place and there are no demagogues;
but where the laws are not supreme, there demagogues spring up. For the
people becomes a monarch and is many in one; and the many have the power
in their hands, not as individuals but collectively.... And the people,
who is now a monarch, and no longer under the control of law, seeks to
exercise monarchical sway, and grows into a despot; the flatterer is
held in honour; this sort of democracy being relatively to other
democracies what tyranny is to other forms of monarchy.

"The spirit of both is the same, and they alike exercise a despotic rule
over the better citizens. The decrees of the Demos correspond to the
edicts of the tyrant, and the demagogue is to the one what the flatterer
is to the other. Both have great power--the flatterer with the tyrant,
the demagogue with democracies of the kind which we are describing. The
demagogues make the decrees of the people override the laws, and refer
all things to the popular assembly. And therefore they grow great,
because the people has all things in its hands and they hold in their
hands the votes of the people, who is too ready to listen to them. Such
a democracy is fairly open to the objection that it is not a
constitution at all; for _where the laws have no authority there is no
constitution_. The law ought to be supreme over all. So that if
democracy be a real form of government, _the sort of constitution in
which all things are regulated by decrees is clearly not a democracy in
the true sense of the word_, for decrees relate only to particulars."

This distinction between true law, that is to say, venerable law, framed
to endure, part of a co-ordinate scheme of legislation, and an emergency
law which is merely a decree like the wishes of a tyrant, constitutes
the whole difference, if we could realise it, between the sociologists
of antiquity and those of to-day. By the term Law, the ancient and the
modern sociologists mean two different things and this is the reason for
so many misunderstandings. When he speaks of law, the modern sociologist
means the expression of the general will at such and such a date, 1910
for instance. The ancient sociologist would consider that the expression
of the general will in the second year of the 73rd Olympiad was not law
at all, but a decree. A law to him would be a paragraph of the
legislation of Solon, Lycurgus or Charondas. Whenever in a Greek or
Roman political treatise we meet the expression--"a State governed by
laws," the only way to translate it is--"a State governed by a very
ancient and immutable legislation." This gives the true meaning to the
famous personification of laws in the Phædo, which would be quite
meaningless if the Greeks had understood what we do by the term. Are
laws the expression of the general will of the people? If so why should
Socrates have respected them, he who despised the people to the day he
was condemned? It would be absurd. These laws which Socrates respected
were not the decrees of the people contemporary with Socrates; they were
the ancient gods of the city, which had protected it from the earliest
days.

These laws may err in that they seemed to sanction the verdict that
condemned Socrates to death, but they were honourable, venerable and
inviolate, because they had been the guardians of the city for
centuries, and guardians of Socrates himself until the day when they
were misapplied against him.

A "constitution," therefore, to adopt Aristotle's terminology, is a
State which obeys laws, that is to say, laws framed by its ancestors.

It is, then, an aristocracy, for it is even more aristocratic to obey
our ancestors themselves by obeying the thoughts which they embedded in
legislation, five centuries ago, than to obey the inheritors of their
tradition, the aristocrats of to-day. For aristocrats of to-day belong
only partly to tradition, in that they live in the present. Whereas a
fifteenth century law belongs to the fifteenth century and to no other
period. To obey law as understood by the ancient sociologists, did not
mean obeying Scipio who has just passed us on the _Via Sacra_. It meant
to obey his grandfather's great grandfather! All this is
ultra-aristocratic.

Precisely! _Law is an aristocratic thing;_ only _the emergency law_, the
_decree_, is democratic. For this reason Montesquieu always speaks of a
monarchy as being limited, and, at the same time, maintained by its law.
What did this mean in his day, when there was no "expression of the
general will" to limit monarchy, and when royalty possessed legislative
power, and could at will make and remake laws? It could only mean one
thing, namely, that Montesquieu's conception of law was the same as that
of the ancient sociologists,--law far older than his time, "fundamental
laws" as he calls them, of the ancient monarchy, which still bind and
ought so to bind the monarch, whose rule without them would be despotism
or anarchy. Law is essentially aristocratic. It ordains that rulers
should govern the people, and that the dead should govern the rulers.
The very essence of aristocracy is the rule of those who have lived over
those who live, for the benefit of those who shall live hereafter.
Aristocracy, properly so called, is an aristocracy in the flesh. Law is
a spiritual aristocracy. Aristocracy, as represented by the aristocrats
of to-day, only represents the dead by tradition, inheritance,
education, physiological heredity of temperament and characteristics.
Law does not represent the dead, it is the dead themselves, it is their
very thought perpetuated in immutable script.

A nation is aristocratic both in form and spirit which preserves its old
aristocracy and maintains its vitality by careful infusions of new
blood. Still more is that nation aristocratic which maintains its old
legislation inviolate, adding to it, reverently and discreetly, new laws
which combine something of the modern spirit with the spirit of the old.
_Homines novi, novæ res. Homo novus_ means the man without ancestors who
is worthy to be added to the ranks of the nobly born. _Novæ res_ are
things without antecedents, nay revolution itself. _Novæ res_ should
only be introduced partially gradually, insensibly and progressively
into ancient things, as "new men" into the community of the old
nobility. Law is more aristocratic than aristocracy itself, hence
democracy is the natural enemy of laws and can only tolerate decrees.

Our examination of modern democracy has brought us to the following
conclusions. The representation of the country is reserved for the
incompetent and also for those biassed by passion, who are doubly
incompetent. The representatives of the people want to do everything
themselves. They do everything badly and infect the government and the
administration with their passion and incompetence.


[B] See _France_, by J. E. C. Bodley, 1899, pp. 334, 335. Under
_Scrutin de liste_ "the department is the electoral unit, each having
its complement of deputies allotted to it in proportion to its
population, and each elector having as many votes as there are seats
ascribed to his department, without, however, the power to cumulate."
_Scrutin d'arrondissement_ is election by single-member constituencies.
The _arrondissement_ is the electoral unit.

[C] This is a question put to a minister by a deputy. "The
effect ... is somewhat similar to a motion to adjourn the house in the
English Parliament." Bodley, p. 445.




CHAPTER VI.

THE INCOMPETENCE OF GOVERNMENT.


This is not all. The law of incompetence spreads still further, either
by some process of logical necessity or by a sort of contagion. It has
often been made the subject of merriment, for, like all tragedy, when we
regard it with good humour the matter has its comic side, that it is
very rare for any high office to be given to a man who is competent for
the post. Generally the Minister of Education is a lawyer; the Minister
of Commerce, an author; the War Minister, a doctor; the Minister for the
Navy, a journalist. Beaumarchais' epigram "The post required a
mathematician--it was given to a dancing master!" strikes the keynote
much more of a democracy than of an absolute monarchy.

The matter is so generally recognised that it has a sort of retroactive
effect upon the historical ideas of the masses. Three Frenchmen out of
every four are convinced that Carnot was a civilian, and the statement
has often appeared in print. Why? because it is inconceivable that under
a democracy the War Minister could possibly be a soldier, or, that the
members of the Convention could possibly have given the War Office to a
soldier. This appeared too paradoxical to be true.

At first sight this extraordinary method of making incompetent men into
ministers seems merely a joke, merely the subtle and entertaining
vagaries of the goddess Incompetence. Partly it is so but not entirely.
The man whose business it is to appoint ministers has to divide the
choicest plums of office among the various groups of the majority which
supports him. As all of these groups do not contain specialists, the
highest offices are disposed of on political grounds, and not on grounds
of professional aptitude. I have shown what the result is; the only
ministerial appointment which is made in a rational manner is that which
the President of the Council reserves for himself, and even in this case
in order to conciliate some important political personage he very often
gives it up and takes some post for which he is not so well suited.

See what follows: each department is directed by an incompetent man,
who, if he be conscientious, sets himself to learn the work in which he
ought to be a fully trained expert, or, if he be not conscientious, and
be pressed for time, as he always is, he directs his department
according to his general political theories and not according to
practical common sense--a double distillation of incompetence.

We know the kind of speech a new Minister of Agriculture makes to his
staff. He harangues them on the principles of the revolution of 1789.

Moreover, in a highly centralised country, the minister does everything
in his own department. He has to do everything under the pressure, it is
true, of the national representatives; but still his is the supreme
authority. It is easy to see what sort of decisions he will make. They
are often very little supported by law, and sometimes are even contrary
to law, and then they remain a dead letter from the first. Ministerial
circulars often have a remarkable character for illegality. In that case
they fall and are forgotten, but not always before they have introduced
a vast amount of trouble throughout the entire administration.

As to appointments, they are made, as I have said, by political
influence, and even when they are flagrantly improper and corrupt, there
is no chance of their being corrected by the competence of a minister,
who, holding enlightened views on the business and subordinates of his
office, is able to put his foot down and say "No! this will not do, we
must draw the line somewhere."




CHAPTER VII.

JUDICIAL INCOMPETENCE.


Here we find incompetence spreading its influence by the logical
necessity of the case. There are other quarters in which it grows by a
sort of contagion. Have you ever noticed that the _ancien régime_, in
spite of grievous shortcomings, by a sort of historical tradition,
maintained a certain respect for efficiency in its different forms? For
instance in matters of jurisdiction, there were seignorial,
ecclesiastical and military courts. These were not founded as the result
of argument and profound consideration, but by the natural course of
events, by history itself, and they were maintained and approved by a
monarchy which was verging on despotism.

Seignorial jurisdiction, without much rational justification, was none
the less of considerable utility; it bound, or was capable of binding,
the noble to his land, it prevented him from losing sight of his
vassals, and his vassals from losing sight of him, and was in fact a
conservative force in the aristocratic constitution of the kingdom. I
submit that if this jurisdiction had been properly defined, limited and
modified, which was never done, it would have been consonant with the
law of competence. There are various local matters which come quite
properly within the province of the noble, who in those days took the
place of the magistrate. All that was wanted was that such matters
should have been defined with precision and that in every case appeal
should have been allowed.

Ecclesiastical jurisdiction was perfectly reasonable, as offences
committed by ecclesiastics have a special character of which
ecclesiastics alone can judge. This seems strange to modern ideas,
although nowadays there are commercial courts and conciliation boards,
because litigation between men of business, between workmen and women
workers, and between employers and employed, can only be decided by men
who have technical knowledge of the subject in dispute. Appeal,
moreover, to a higher court is always allowed.

Finally, in the old days there used to be military jurisdiction for
precisely the same reason.

All these exceptional jurisdictions are objects of the liveliest
apprehension to democracy, because they infringe the rule of uniformity,
which is the image and often the caricature of equality, and also
because they are a stronghold of efficiency.

Democracy of course demolished aristocratic courts together with the
aristocracy itself, and ecclesiastical courts together with the Church
when it ceased to be an estate of the realm. Any special jurisdictions
which still remain are looked upon as instruments of aristocracy;
courts-martial are held in abhorrence because they have ideas of their
own in respect of military honour and duty, and military offences.
Therein lies their efficiency, a thing absolutely necessary, if we are
to maintain military spirit and discipline in a strong army. The private
soldier or officer, who is only judged and punished as a civilian, will
not be well judged nor adequately punished, considering the special
duties and services which are required of the army. This is a question
of moral as well as technical efficiency and to this the democracy pays
no heed, because it is convinced that no special efficiency is necessary
and that common sense is all that is required. Common sense, however, is
like wit; it is useful in every walk of life, but is not sufficient in
any one of them. This is just what democracy cannot or will not
understand.

It makes just as great a mistake in its civil and criminal jurisdiction,
though it has, up to now, so far departed from its principles as to
appoint qualified jurists to civil judgeships. No one denies that this
body of men is efficient. Those who act as judges know their law. There
is, however, as I have often had occasion to point out, a moral as well
as a technical efficiency, and in limiting the independence that is
essential to moral efficiency, democracy neutralises the technical
efficiency of its servants. Let me explain my meaning further.

Formerly the magistracy was a recognised and autonomous branch of the
public service, and as a result, save as it was affected by revolution
and in normal times by the fear of revolution, enjoyed an absolute
independence. This gave, or rather preserved intact, its moral
efficiency. For moral efficiency consists in an ability to act
according to the dictates of conscience, and is equivalent to a sort of
moral independence.

Now, the magistrates form a department of the administration and are a
body of officials. The State appoints, promotes or refuses to promote
and pays them. In short the State has them at its mercy, just as
military officers are controlled by the War Office, or tax-collectors by
the Treasury. Hence they are deprived of their independence and moral
efficiency, for they are always tempted to give judgment as the
Government would wish.

There is, it is true, a guarantee for their independence in the
permanence of their appointments, but this only applies to those who
have reached the summit of their profession, or are on the point of
retiring, or have no further interest in promotion. The young magistrate
who wants to get on, a perfectly legitimate ambition, is by no means
independent, for if he does not give satisfaction, he may enjoy a
peculiar kind of permanence, the permanence of standing still at the
starting point. The only independent judges, to whom justice is the sole
interest, are either those who have served for forty years or the
President of the _Cour de Cassation_. I may add also the man of
independent means who is indifferent to promotion and content to spend
all his time at the place of his first appointment. He is exactly like
the magistrates in old days, but he and his kind get rarer every year.

At best, moreover, this permanence, of which so much is thought, is an
illusory guarantee, for it is often suspended by one Government or
another, and the magistrates are constantly at the mercy of political
crises. Their moral efficiency is indeed sorely tried.

I affirm, therefore, that this diminution of moral efficiency affects
technical efficiency, because magistrates dare not insist on technical
exactitude when cases arise between the State and individuals, or
between those who are protected by Government and those who are not.
Though cases in which the State is a party do not occur very often,
those in which friends of the Government are involved are of daily
occurrence in a country where Government is a faction waging incessant
warfare against all other factions.

It has been said with much reason that parliamentary government on a
basis of universal suffrage is legalised and continuous civil war. It is
usually a bloodless civil war, but its weapons are insults,
provocations, calumnies, personalities, libel actions. These go on from
one year's end to the other. In a country where such a state of affairs
is prevalent, the magistracy ought to be absolutely independent in order
to be impartial. Yet it is precisely in a country like this that the
magistracy, not being independent and autonomous, is obliged to avoid
offending the party in office which, moreover, is extremely exacting,
for it lives in constant fear that it may be turned out of power.

--Is there nothing to be done? Would you advocate a return to the
practice of purchasing judicial appointments?--

In the first place, this would not be anything so very terrible, and
secondly, it might be quite possible to secure all the advantages of
purchase without its actual practice.

I can show you that it is not so very terrible, for the case is parallel
with that of the exceptional jurisdictions, the mention of which filled
you with horror till you remembered the commercial courts and the
councils of experts, all excellent institutions. We are appalled at the
idea of a magistrate purchasing his office, and yet we employ advocates
and solicitors and other legal officials and trust them with our most
precious interests, yet they have, many of them, either bought or
inherited their practice. Under a system of purchase, we should be
judged by lawyers of whom we required more extensive legal knowledge
than is at present required of the profession. We should be judged in
fact by solicitors and advocates of a superior order. There is nothing
very alarming about that.

Montesquieu was in favour of a system of purchase. Voltaire opposed it
strongly. They were both right and were indeed agreed on general
principles. Montesquieu says: "Venality,--the purchase system,--is a
good thing under a monarchical form of government, because work which
would not be done from mere civic virtue is then undertaken as a family
business. Each man's duty is laid down for him, and the orders of the
State are given greater permanence. Suidas says very aptly of Anastasius
that he turned the Empire into an aristocracy by selling magisterial
offices."

Voltaire replies: "Is it as a matter of civic virtue that in England a
judge of the King's Bench accepts his appointment?" (It is either a
matter of civic virtue or of profit and interest, and if it is not
profit, it certainly must require considerable civic virtue.) "What! can
we not find men in France willing to judge if we bestow their
appointments upon them gratuitously?" (We certainly can: but they might
be too grateful!) "Can the work of administering justice, disposing of
the lives and fortunes of men, become a family business?" (Well, the
business of bearing arms and disposing of men's lives and fortunes in
civil war was in 1760 a family business. So too the business of being
king, and you do not protest against that!) "It is a pity that
Montesquieu should dishonour his work by such paradoxes, but we must
forgive him; his uncle purchased a provincial magistrate's office and
left it to him. Human nature comes in everywhere. None of us is without
weaknesses."

Montesquieu thinks aristocratic bodies are good things. Voltaire is in
favour of absolute power. Montesquieu would like the judicature to be a
family office, that is to say hereditary like the profession of a
soldier; this would make the judicial profession permanent like other
professions. He demonstrates, as does Suidas, that the purchase system
creates an aristocracy. Voltaire, like Napoleon I., would make his
soldiers, his priests, and his judges, king's men. They should all
belong to the king, body and soul.

Montesquieu had a greater antagonist than Voltaire in Plato. Plato wrote
in his Republic, referring to all judicial offices: "It is as if on
board ship a man were made a pilot for his wealth. Can it be that such a
rule is bad in every other calling, and good only in respect of the
governing of a republic?"

Montesquieu answers Plato (and in anticipation Voltaire) very wittily:
"Plato is speaking of a virtuous republic and I of a mere monarchy.
Under a monarchy if offices were not sold by rule, the poverty and greed
of courtiers would sell them all the same, and chance after all will
give a better result than the choice of a prince."

To sum up, Montesquieu wants the magistracy to be partly hereditary, and
partly recruited from the wealthy classes, an independent, aristocratic
body analogous to the army or the clergy, administering justice with
that technical efficiency which university standards can guarantee, and
with the moral efficiency which is founded on independence, dignity,
public spirit and impartiality.

I said above that venality, or the system of purchase, was not necessary
to obtain these results. The principle is this, that the magistracy must
be independent, and to be independent it must have a proprietary right
in its duties. This can only be obtained if it hold its office by
inheritance or purchase as was done under the _ancien régime_; or, if it
were somehow contrived that magistrates should not be chosen by the
Government. The purchase or inheritance plan is not popular, then the
only alternative is that the magistrates should be chosen by some body
other than the Government. By whom then? The people? Then the judges
would be dependent upon the people and the electors.

--That would be better, or less bad.--

Not at all. If the judges were chosen by the electors, they would be
even less impartial than if they were elected by the Government. The
judge then would think of nothing but of being re-elected. He would
always give judgment in favour of the party which had elected him.
Would you care to be judged before a court composed of the deputies of
your department? Certainly not, if you belong to the weaker party. Yes,
if you belong to the majority, but then only if you are certain that
your adversary belongs to the minority, or, if he belong to your own
party, that he is a less influential elector than yourself. To sum up,
there is no guarantee of impartiality if the judges are elected.

Further, if the system of electing judges by those liable to their
jurisdiction were adopted, there would be an extensive and, I might add,
a most entertaining variety of justice. Judges, who were elected by a
"blue" or republican majority, and who were anxious for re-election,
would always deliver judgment in favour of the blues. The same thing
would happen in the "white" or royalists districts. "Justice has her
epochs," Pascal said ironically, and in this case justice would have her
districts. It would not be the same in the _Alpes-Maritimes_ as in the
_Côtes-du-Nord_. The Court of Appeal, if it attempted to be impartial,
would spend its time sending cases back from a blue district to be
revised in a white, and the decisions delivered in a white country to
be revised in a blue. There would be judicial and legal anarchy.

--If the bench is not to be inherited, nor bought, nor chosen by the
Government, nor elected by the people, by whom is it to be nominated?--

By itself; I see no other solution.

For instance I can suggest one good method, though there may be several.
All the doctors of law in France could choose the judges of appeal and
the judges of appeal could choose and promote all the judges. This is an
aristocratic-democratic scheme on a very broad basis.

Or else the judges alone might choose the judges of appeal, and the
judges of appeal might appoint and promote the judges. That is an
oligarchical method.

Or again, here is a plan for passing from the system that is, to that
which ought to be. For the first time the doctors of law might choose
the _Cour de Cassation_, and it could choose the judges. Afterwards the
judges could fill the vacancies in the _Cour de Cassation_, which would
nominate and promote the judges.

The Government would still go on, and continue to nominate the persons
eligible to serve as magistrates.

Under all these systems the judges would form an autonomous,
self-creative body, dependent upon and responsible to themselves alone,
and by reason of their absolute independence, strictly impartial.

--But they would form a caste!--

They would form a caste. I am sorry for it, but it is the case. You will
never be well judged until you have a judicial caste, which is neither
the Government, nor the world at large. For the Government cannot judge
properly when it is both judge and party to the suit. Further, if it be
litigious; it will never be out of court. Again, the world at large
cannot judge properly, because, in practice, the world at large means
the majority, and the majority is a party, and by definition a party can
hardly be impartial.

But democracy does not want to be judged by a caste. In the first place
because it abhors castes, and secondly because it does not care about
impartial justice. Do not exclaim at the paradox. Democracy does want to
be judged impartially in little every-day cases, but in all important
cases in which a political question is involved and in which one of the
majority is opposed to one of the minority, the verdict then has to be
for the stronger side.

It says to the judicial bench what a simple-minded deputy said to the
President of the Chamber: "It is your duty to protect the majority."

This is why democracy clings to its official magistracy, which contains
some good elements though its members cannot always be impartial. They
were condemned by the mouth of one of their highest dignitaries who
answered when questioned about some illegal proceeding: "There are
reasons of high State policy," thus throwing both the law and the judges
at the feet of the Government. On another occasion, with the very best
intentions, in order to put an end to an interminable affair, they
turned and twisted the law and set a bad example; for by not applying
the law correctly, they laid themselves open to endless and justifiable
attacks upon their decision; they did not procure the longed-for
settlement, and, instead, left the matter open to interminable dispute.
They have knowledge, good sense and intelligence, but as their want of
independence, in other words their moral inefficiency, neutralises
their technical efficiency, they do not and cannot possess authority.

Democracy will inevitably go further along the road towards its ideal,
which is direct government. It will want to elect the judges.

Already it chooses them remotely in the third degree; for it chooses the
deputies who choose the Government, which chooses the judges; and to
some extent, in the second degree, for it chooses the deputies who bring
pressure to bear upon the nomination of the judges and interfere with
their promotion and their decisions. This also is remote.

And, as by this constitution, or, rather by this practice, recognition
is given to the principle that it is the people who really appoints the
judges through its intermediaries, democracy, always logical and matter
of fact, would like to see the principle applied without concealment,
and the people making the appointments directly.

Then endless questions will arise about the best way of voting and
electing. If unipersonal ballot is adopted, the canton will nominate its
_juge de paix_, the district its tribunal, the region its Court, and the
whole country the Court of Appeal. In this arrangement there will be
the double drawback mentioned above; that is, varying interpretations of
justice according to districts, and no impartiality.

If, on the other hand, _scrutin de liste_ is adopted, the whole country
will choose all the magistrates and they will belong to the majority. In
this case there would be uniformity of justice but no impartiality. Any
intermediate system would combine the disadvantage of both plans. For
instance, if nominations are made in each division, all the magistrates
in Brittany will be white partisans, while in Provence they will be blue
partisans. In both cases they will be biassed, and such diversity as
there is will be merely a diversity of partiality and bias.

We are talking of the future, though not perhaps of a very distant one.
Let us deal with the present. The jury is still with us. Now the jury
combines absolute moral competence with absolute technical incompetence.
Democracy must always have incompetence in one form or another. A jury
is independent of everybody, both of the Government and of the people,
and in the best possible way, because it is the agent of the people
without being elected. It does not seek re-election and is rather vexed
than otherwise at being summoned to perform a disagreeable duty. On the
other hand it always vacillates between two emotions, between pity and
self-preservation, between feelings of humanity and the necessity for
social protection; it is equally sensitive to the eloquence of the
defending advocate, and the summing up of the prosecutor, and as these
two influences balance each other it is in a perfect moral condition for
delivering an equitable verdict.

For this reason the jury is of ancient origin, and has always been an
institution in the land. At Athens the tribunal of the Heliasts formed a
kind of jury, too numerous indeed and more like a public meeting, but
still a sort of jury.

At Rome, a better regulated republic, there were certain citizens chosen
by the prætor who settled questions of fact, that is to say, decided
whether an act had or had not been committed, whether a sum of money had
or had not been paid; and the question of law was reserved for the
centumvirs.

In England the jury still exists and has existed for centuries.

These various peoples have considered very properly that juries are
excellently adapted for forming equitable decisions, since they possess
a greater moral competence for this particular function, than is to be
found elsewhere.

This is true; but on the other hand a jury has no intelligence. In
November 1909, a jury in the Côte d'Or before whom a murderer was being
tried, declared (1) that this man did not strike the blows, (2) that the
blows which he struck resulted in death. Thereupon the man was
acquitted, although his violence, which never took place, had a
murderous result.

In the Steinheil case in the same month and year, the jury's verdict
involved (1) that no one had been assassinated in the Steinheils' house,
and (2) that Mme Steinheil was not the daughter of Mme Japy. If a
verdict were a judgment this would have put an end to all attempts to
discover the assassins of M. Steinheil and Mme Japy, and on the other
hand there would have been terrible social complications.

But the verdict of a jury is not a judgment. Why? Because the legislator
foresaw the alarming absurdity of verdicts. It is presumed in law that
all juries' verdicts are absurd, and experience proves that this is
often the case. Juries' verdicts always seem to have been decided by lot
like those of the famous judge in Rabelais, and it is proverbial at the
law courts that it is impossible to foresee the issue of any case that
comes before a jury. It looks as if the jury reasoned thus: "I am a
chance judge, and it is only right that my judgment should be dictated
by chance."

Voltaire was in favour of the jury system, principally because he had
such a very low opinion of the magistrates of his day, whom he used to
compare to Busiris. But, with his usual inconsequence, he takes no pains
to conceal the fact that the populations of Abbeville and its
neighbourhood were unanimously exasperated against La Barre and
D'Etalonde, and the people of Toulouse against Calas, and all of them
would have been condemned by juries summoned from those districts as
surely as they were by the magisterial Busiris.

The jury system is nothing but a refined example of the cult of
incompetence. Society, having to defend itself against thieves and
murderers, lays the duty of defending it on some of its citizens, and
arms them with the weapon of the law. Unfortunately it chooses for the
purpose citizens who do not know how to use the weapon. It then fondly
imagines that it is adequately protected. The jury is like an unskilled
gladiator entangled in the meshes of his own net.

I need hardly say that democracy with its usual pertinacity is now
trying to reduce the jury a step lower, and draw it from the lower
instead of the lower middle classes. I see no harm in this myself, for
in the matter of law the ignorance and inexperience of the lower middle
class and the ignorance of the working class are much the same. I have
only mentioned it to show the tendency of democracy towards what is
presumably greater incompetence.

Now comes the turn of the _juges de paix_. At present we still have
_juges de paix_. Here we have a most interesting example of the way
democracy strives after incompetence in matters judicial.

Owing to the expense entailed by an appeal the jurisdiction of a _juge
de paix_ is very often final. He ought to be an instructed person with
some knowledge of law and jurisprudence. He is therefore usually chosen
from men who have a degree in law or from lawyers' clerks who have a
certificate of ability. To be quite honest this is but a feeble
guarantee.

By the law of July 12th, 1905, the French Senate, anxious to find men of
still grosser incompetence, decided that _juges de paix_ might be
nominated from those, who, not having the required degree or
certificate, had occupied the posts of mayor, deputy-mayor or councillor
for ten years.

The object of this decision was the very honest and legitimate one of
giving senators and deputies the opportunity of rewarding the electoral
services of the village mayors and their assistants. And remember
senators especially are nominated by these officials. Further it was an
opportunity not to be missed for applying our principle--and our
principle is this: we ask, where is absolute incompetence to be found,
for to him who can lay indisputable claim to it we must confide
authority.

Now mayors and their assistants answer this description exactly. They
must be able to sign their names, but they are not obliged to know how
to read, and eighty per cent. of them are totally illiterate. Their work
is done for them very usually by the local schoolmaster. The Senate,
therefore, was quite sure of finding among them men absolutely
incompetent for the post of _juge de paix_, and it has found what it
wanted. Incompetence so colossal deserved an appointment, and an
appointment has been given to it.

The magistrature and the powers that be, seem to have been somewhat
disturbed by certain consequences of this highly democratic institution.
M. Barthou, the Minister of Justice, complained bitterly of the work
which this new institution caused him. He made the following speech in
the Chamber of Deputies: "We are here to tell each other the truth, and,
with all the due moderation and prudence that is fitting, I feel it my
duty to warn the chamber against the results of the law of 1905. At the
present moment I am besieged with applications for the post of _juge de
paix_. I need hardly mention that there are some 9,000 of them in my
office, because a certain number are not eligible for consideration, but
there are in round numbers 5,500 applications which are recommended and
examined." (What he means to say is, that these are examined because
they have been recommended, for, as is only right, those that are not
backed by some political personage are not looked at.) "As the average
annual number of vacancies is a hundred and eighty, you will readily see
what a quandary I am in. Some of these applications are made with the
most extraordinary persistency, I might even call it ferocity, and these
invariably come from men who have held the office of mayor or
deputy-mayor for ten years, often in the most insignificant places."

The Minister of Justice then read a report made on the subject by a
_procureur-général_.

"In this department there are forty-seven _juges de paix_, twenty of
whom, as I learn from an enquiry, were mayors at the time of their
appointment. It is not to be wondered at that the number of provincial
magnates who aspire to the post is on the increase, for it seems to be
generally recognised in this department that elective office
irrespective of all professional aptitude is the normal means of access
to a paid appointment, more especially to that of _juge de paix_. Once
they are appointed, the mayors combine both their municipal and judicial
duties, and their interests lie far more in the commune which they
administer than in the district in which they dispense justice and
which, without permission, they should never leave. Sometimes these
district magistrates will go to any length to obtain moral support from
the politicians of the neighbourhood. They extort this as a sort of
blackmail given in exchange for the electoral influence which they can
bring to bear in their municipal capacity. They attach far less
importance to being quashed by the bench, than to the eventual support
of the deputy. Those who come into their courts are the unfortunate
victims of these compromising arrangements which are giving the
Republican system a bad name."

I think the Minister of Justice and his _procureur-général_ have very
little ground for these lamentations. After all the minister only
complains of having 9,000 applications for office. It would surely be
quite easy for him, in compliance with the generally recognised
principle, to choose those whose incompetence seems to be most thorough,
or those who are most influentially supported, according to the
prevailing custom.

As for the _procureur-général's_ sarcasms, which he thinks so witty,
they are quite delightfully diverting and ingenuous. "It seems to be
generally recognised that elective office, irrespective of all
professional aptitude, is the normal means of access to a paid
appointment." What else does he expect? It is eminently democratic that
the marked absence of professional capacity should single a man out for
employment. That is the very spirit of democracy. He surely does not
think that a man is an elector by reason of his legislative and
administrative capacity?

It is likewise essentially democratic that elective office should lead
to paid appointments, for the democratic theory is that all office, paid
and unpaid, should be elective. Why, this _procureur-général_ must be an
aristocrat!

As for the mutual services rendered by the justice, as mayor, to the
deputy, and by the deputy to the justice, this is democracy pure and
simple. The deputies distribute favours that they may be returned to
power; the influential electors put all their interest, both personal
and official, at the service of the deputies in order to obtain those
favours. They are hand in glove with each other, and form a solid
union of interests.

What more does the _procureur-général_ want? Does he want a different
system? If he wants another system, whatever else it may be, it will not
be democracy, or at least it will not be a democratic democracy. Nor
have I any idea what he means when he says the Republican system will
get a bad name. The good name of the Republic depends upon its putting
into practice every democratic principle; and democratic principles have
certainly never been more precisely realised than in the preceding
example, which I have had great pleasure in rescuing from oblivion and
presenting to the notice of sociologists.




CHAPTER VIII.

EXAMPLES OF INCOMPETENCE.


I have already compared this, our desire to worship incompetence, to an
infectious disease. It has attacked the State at the very core, in its
constitution, and it is not surprising that it is spreading rapidly to
the customs and to the morals of the country.

The stage, we know, is an imitation of life. Life also, to perhaps an
even greater extent, is an imitation of the stage. Similarly laws spring
from morals, and morals spring from law. "Men are governed by many
things," said Montesquieu, "by climate, religion, laws, precept,
example, morals and manners, which act and react upon each other and all
combine to form a general temperament."

Morals, more often than not, determine the nature of our laws,
particularly in a democracy, which is deplorable, but Montesquieu was
right in saying: "Morals take their colour from laws, and manners from
morals," for laws certainly "help to form morals, manners" and even
"national character." For instance in Rome under the Empire the code of
morals was to some extent the result of arbitrary power, as to-day the
moral character of the English is to some extent due to the laws and
constitution of their country.

We know that by his laws Peter the Great changed if not the character at
least the manners and customs of his people.

Custom is the offspring of law, and morals are the offspring of custom.
National character is not really changed, for character, I believe, is a
thing incapable of change, but it appears to be changed, and it
certainly undergoes some modifications; one set of tendencies is
checked, while others are encouraged.

The law abolishing the right of primogeniture has obviously affected
national morals, though it has not otherwise altered national character.
For a peculiar mental attitude is evolved by the constant domination of
an elder brother, whose birthright gives him precedence and authority
second only to that of the father. In countries where the right of
unrestricted testamentary bequests is still maintained, family morals
are very different from those which obtain where the child is
considered a joint proprietor of the patrimony.

Since the passing of the law permitting divorce, a sad but necessary
evil, there have been far more applications for divorce than there ever
were for separation. Can this be accounted for solely by the fact that
formerly it seemed hardly worth while to take steps to obtain the
qualified freedom of separation? I think not. For when a yoke is
unbearable, efforts to relax it would naturally be quite as strenuous
and as unremitting as efforts to get rid of it altogether.

The truth is, I think, that when both civil and ecclesiastical law
agreed in prohibiting divorce, people held a different view of marriage;
it was looked upon as something sacred, as a tie that it was shameful to
break, and that could not be broken except as a last resource and then
almost under pain of death. The law permitting divorce was what our
forefathers would have called a "legal indiscretion." It has abolished
the feeling of shame. Except where there is strong religious feeling,
there is now no scruple nor shame in seeking divorce. The old order has
passed away; modesty has been superseded by a desire for liberty, or for
another union. This change has been brought about by a law which was the
result of a new moral code; but the law itself has helped to enlarge and
expand the code.

Thus democracy extends that love of incompetence which is its most
imperious characteristic. Greek philosophers used to delight in
imagining what morals, especially domestic morals, would be like under a
democracy. They all vied with Aristophanes. One of Xenophon's characters
says: "I am pleased with myself, because I am poor. When I was rich I
had to pay court to my calumniators, who knew full well that they could
harm me more than I could them. Then the Republic was always imposing
fresh taxes and I could not escape. Now that I am poor, I am invested
with authority; no one threatens me. I threaten others. I am free to
come and go as I choose. The rich rise at my approach and give me place.
I was a slave, now I am a king; I used to pay tribute, now the State
feeds me. I no longer fear misfortunes, and I hope to acquire wealth."

Plato too is quietly humorous at democracy's expense. "This form of
government certainly seems the most beautiful of all, and the great
variety of types has an excellent effect. At first sight does it not
appear a privilege most delightful and convenient that we cannot be
forced to accept any public office however eligible we may be, that we
need not submit to authority and that every one of us can become a judge
or magistrate as our fancy dictates? Is there not something delightful
in the benevolence shown to criminals? Have you ever noticed how, in
such a State as this, men condemned to death or exile remain in the
country and walk abroad with the demeanour of heroes? See with what
condescension and tolerance democrats despise the maxims which we have
been brought up from childhood to revere and associate with the welfare
of the Republic. We believe that unless a man is born virtuous, he will
never acquire virtue, unless he has always lived in an environment of
honesty and probity and given it his earnest attention. See with what
contempt democrats trample these doctrines under foot and never stop to
ask what training a man has had for public office. On the contrary,
anyone who merely professes zeal in the public interests is welcomed
with open arms. It is instantly assumed that he is quite disinterested.

"These are only a few of the many advantages of democracy. It is a
pleasant form of government _in which equality reigns among unequal as
well as among equal things_. Moreover, when a democratic State, athirst
for liberty, is controlled by unprincipled cupbearers, who give it to
drink of the pure wine of liberty and allow it to drink till it is
drunken, then if its rulers do not show themselves complaisant and allow
it to drink its fill, they are accused and overthrown under the pretext
that they are traitors aspiring to an oligarchy; for the people prides
itself on and loves the equality that confuses and will not distinguish
between those who should rule and those who should obey. Is it any
wonder that the spirit of licence, insubordination, and anarchy should
invade everything, even the institution of the family? Fathers learn to
treat their children as equals and are half afraid of them, while
children neither fear nor respect their parents. All the citizens and
residents and even strangers aspire to equal rights of citizenship.

"Masters stand in awe of their disciples and treat them with the
greatest consideration and are jeered at for their pains. Young men want
to be on the same terms as their elders and betters, and old men ape the
manners of the young, for fear of being thought morose and dictatorial.
Observe too to what lengths of liberty and equality the relations
between the sexes are carried. You would hardly believe how much freer
domestic animals are there than elsewhere. It is proverbial that little
lap-dogs are on the same footing as their mistresses, or as horses and
asses; they walk about with their noses in the air and get out of
nobody's way."

Aristotle, faithless at this point to his favourite method of always
contradicting Plato, has no particular liking, as we have said, for
democracy. He does not spare it though he does not imitate Plato's
scathing sarcasm.

In the first place, Aristotle is frankly in favour of slavery, as was
every ancient philosopher except perhaps Seneca; but he is more
insistent on this point than anyone else, for he looks upon slavery, not
as one of many foundations, but as the very foundation of society.

He considers artisans as belonging to a higher estate but still as a
class of "half-slaves." He asserts as an historical fact that only
extreme and decadent democracies gave them rights of citizenship, and
theoretically he maintains that no sound government would give them the
franchise of the city. "Hence in ancient times, and among some nations,
the working classes had no share in the government--a privilege which
they only acquired under the extreme democracy.... Doubtless in ancient
times and among some nations the artisan class were slaves or
foreigners, and therefore the majority of them are so now. The best form
of State will not admit them to citizenship...."

He admits that democracy may be considered as a form of government ("...
if democracy be a real form of government...."), and he admits too
that "... multitudes, of which each individual is but an ordinary
person, when they meet together, may very likely be better than the few
good, if regarded not individually but collectively.... Hence the many
are better judges than a single man of music and poetry; for some
understand one part, and some another, and among them they understand
the whole. [Observe that he is still speaking of a democracy in which
slaves and artisans are not citizens.] Doubtless too democracy is the
most tolerable of perverted governments, and Plato has already made
these distinctions, but his point of view is not the same as mine. For
he lays down the principle that of all good constitutions democracy is
the worst, but the best of bad ones." But still Aristotle cannot help
thinking that democracy is a sociological mistake "... It must be
admitted that we cannot raise to the rank of citizens all those, even
the most useful, who are necessary to the existence of the State."

Democracy has this drawback that it cannot constitutionally retain
within itself and encourage eminent men. In a democracy "if there be
some one person or more than one, although not enough to make up the
whole complement of a State, whose virtue is so pre-eminent that the
virtues or the capacity of all the rest admit of no comparison with his
or theirs, he or they can be no longer regarded as part of a State; for
justice will not be done to the superior, if he is reckoned only as the
equal of those who are so far inferior to him in virtue and in political
capacity. Such an one may truly be deemed a God among men. Hence we see
that legislation is necessarily concerned only with those who are equal
in birth and in power; and that for men of pre-eminent virtue there is
no law--they are themselves a law. Anyone would be ridiculous who
attempted to make laws for them: they would probably retort what, in the
fable of Antisthenes, the lions said to the hares--'where are your
claws?'--when in the council of the beasts the latter began haranguing
and claiming equality for all. And for this reason democratic States
have instituted ostracism; equality is above all things their aim, and
therefore they ostracise and banish from the city for a time those who
seem to predominate too much through their wealth, or the number of
their friends, or through any other political influence. Mythology tells
us that the Argonauts left Heracles behind for a similar reason; the
ship Argo would not take him because she feared that he would have been
too much for the rest of the crew."

Thrasybulus, the tyrant of Miletus, asked Periander, the tyrant of
Corinth, one of the seven sages of Greece, for advice on the art of
government. Periander made no reply but proceeded to bring a field of
corn to a level by cutting off the tallest ears. "This is a policy not
only expedient for tyrants or in practice confined to them, but equally
necessary in oligarchies and democracies. Ostracism is a measure of the
same kind, which acts by disabling and banishing the most prominent
citizens."

This is what we may call a constitutional necessity for the democracy.

To be quite honest, it is not always obliged to cut off the ears of
corn. It has a simpler method. It can systematically prevent any man who
betrays any superiority whatsoever, either of birth, fortune, virtue or
talent, from obtaining any authority or social responsibility. It can
"send to Coventry." I have often pointed out that under the first
democracy Louis XVI was guillotined for having wished to leave the
country, while under the third democracy his great-nephews were exiled
for wishing to remain in it. Ostracism is, in these instances, still
feeling its way, and its action is contradictory because it has not made
up its mind. This will continue till it has been reduced to a science,
when it will contrive to level, by one method or another, every
individual eminence, great and small, that dares to vary by the merest
fraction from the regulation standards. This is ostracism, and
ostracism, so to speak, is a physiological organ of democracy. Democracy
by using it mutilates the nation, without it democracy would mutilate
itself.

Aristotle often tries to solve the problem of the eminent man. "Good
men," he says, "differ from any individual of the many, as the beautiful
are said to differ from those who are not beautiful, and works of art
from realities, because in them the scattered elements are combined....
Whether this principle can apply to every democracy and to all bodies of
men is not clear.... But there may be bodies of men about whom our
statement is nevertheless true. And if so, the difficulty which has been
already raised--viz., what power should be assigned to the mass of
freemen and citizens--is solved. There is still a danger in allowing
them to share the great offices of State, for their folly will lead them
into error and their dishonesty into crime. But there is a danger also
in not letting them share, for a State in which many poor men are
excluded from office will necessarily be full of enemies. The only way
of escape is to assign to them some deliberative and judicial
functions.... But each individual left to himself, forms an imperfect
judgment."

It is not only the eminent man that is the thorn in the flesh of
democracies, but every form of superiority, whether individual or
collective, which exists outside the State and the Government.

If we recollect that Aristotle coupled extreme democracy with tyranny,
it will be interesting to recall his summary of the "ancient
prescriptions for the preservation of a tyranny...." "The tyrant should
lop off those who are too high; he must put to death men of spirit: he
must not allow common meals, clubs, education and the like; he must be
upon his guard against anything which is likely to inspire either
courage or confidence among his subjects; he must prohibit literary
assemblies or other meetings for discussion, and he must take every
means to prevent people from knowing one another (for acquaintance
begets mutual confidence)." Aristotle's conclusions are subjectively
aristocratic: "In the perfect State there would be great doubts about
the use of ostracism, not when applied to excess in strength, wealth,
popularity or the like, but when used against some one who is
pre-eminent in virtue. What is to be done with him? Mankind will not say
that such an one is to be expelled and exiled; on the other hand he
ought not to be a subject, that would be as if men should claim to rule
over Zeus on the principle of rotation of office. The only alternative
is that all should joyfully obey such a rule, according to what seems to
be the order of nature, and that men like him should be kings in their
State for life." But when he speaks objectively, Aristotle comes to
another conclusion, which we shall have occasion to mention later on.

Among moderns, Rousseau declared that he was not a democrat, and he was
right, because by democracy he meant the Athenian system of direct
government, of which he did not for an instant approve. In the "Social
Contract" he has drawn up a most detailed scheme, which, in spite of
some contradictions and obscure passages, is an exact description of
democracy as we understand the word; but still we cannot tell if he is
actually a democrat, because we do not know what he means by "citizens,"
whether he means everybody or only one class, though that a numerous
one. Rousseau has written more fully than anyone else, not so much of
the influence of democracy on morals, as of the _coincidence_ between
democracy and good morals. Equality, frugality and simplicity can all be
found, according to Rousseau, in States where there is neither royalty
nor aristocracy nor plutocracy. As I understand it, his meaning is that
the same virtue which makes certain nations love equality, frugality and
simplicity is also productive of a form of government which excludes
aristocracy, plutocracy and royalty. If you have simplicity, frugality
and equality, you will probably live in a republic that is democratic or
virtually democratic. This is, I think, the clearest and most impartial
summary that we can make of Rousseau's doctrine, which, though set forth
in rigid formulæ, is still extremely vague.

In this he is a far more faithful follower of Montesquieu than he will
allow. All that I have quoted is to be found literally in Montesquieu's
chapters on democracy. Even his famous saying, "the ruling principle of
democracy is virtue," means, when he uses it in one sense, no more than
that it is the synthesis of these three perfections, equality,
simplicity and frugality. For Montesquieu sometimes uses "virtue" in a
narrow, and sometimes in a broad sense, sometimes in the sense of
political and civic virtue or patriotism, sometimes in the sense of
virtue properly speaking (simplicity, frugality, thrift, equality). In
this latter case he and Rousseau are absolutely agreed.

Montesquieu only considers democracy in decadence, as his custom is in
respect of other forms of government, and though he does not actually
cite Plato, he really gives the substance of what we have already
quoted. "When the people wishes to do the work of the magistrates, the
dignity of the office disappears and when the deliberations of the
Senate carry no weight, neither senators nor old men are treated with
respect. When old men do not receive respect, fathers cannot expect it
from their children, husbands from their wives, nor masters from their
men. At length everyone will learn to rejoice in this untrammelled
liberty, and will grow as weary of commanding as of obeying. Women,
children and slaves will submit to no authority. There will be an end of
morals, no more love of order, no more virtue."

Now as to this transition, this passage from the public morals of a
democracy to the private, domestic, personal morals which exist under
that form of government, have you observed what is the common root of
our failings both public and private? The common root of both is
misunderstanding, forgetfulness and contempt of competence. If pupils
despise their masters, young men despise old men, if wives do not
respect their husbands and the unenfranchised do not respect the
citizens, if the condemned do not stand in awe of their judges, nor sons
in awe of their parents, the principle of efficiency has vanished.
Pupils no longer admit the scientific superiority of their teachers,
young men have no regard for the experience of the old, women will not
recognise the supremacy of their husbands in practical matters, the
unenfranchised have no sense of the superiority of the citizens from the
point of view of national tradition, the condemned do not feel the moral
supremacy of their judges, and sons do not realise the scientific,
practical, civic and moral superiority of their fathers.

Indeed, why should they? How could we expect these feelings to be of
anything but the most transient description since the State itself is
organised on a basis of contempt for competence, or of what is even
worse, a reverence for incompetence, and an insatiable craving for the
guidance and government of the incompetent?

Thus public morals have a great influence on private morals; and
gradually into family and social life there comes that laxity in the
daily relations of the citizens which Plato has wittily termed,
"equality between things that are equal and those that are not."

The first innovation which democracy brings into family life is the
equality of the sexes, and this is followed by woman's disrespect for
man. This idea, be it admitted, is substantially correct, it only ceases
to be true when it is viewed relatively to the varying competences of
the two sexes. Woman is man's equal in cerebral capacity, and in
civilised societies, where intellect is the only thing that matters, the
woman is the equal of the man. She should be admitted to the same
employments as men in society, and under the same conditions of capacity
and education, but in family life the same rules should apply as in
every other enterprise; (1) division of labour according to the
competence of each; (2) recognition of a leader according to the
competence of each. This is the law which women are constantly led to
misunderstand in a democracy. They will not admit the principle of the
division of labour either in the world at large or in the domestic
circle. They try to encroach upon men's work, which perhaps they might
do very successfully, if they were obliged to do it and had nothing else
at all to do; but which they really spoil by undertaking when they have
other obvious duties to perform. They will not admit that men should be
at the head of affairs; they aspire to be not only partners but managing
directors. This implies a contemptuous rejection of that form of social
competence which comes from the acceptance of convention or contract. No
doubt a woman would be just as good a tax-collector as her husband, but
since they have entered into partnership, the one to administer the
collection of taxes, the other to look after the house, it is just as
bad for the one whose business it is to keep house to begin collecting
taxes, as it is for the tax-collector to interfere with the
housekeeping. It is necessary to respect the efficiency that arises out
of the observance of convention and contract. This, with practice and
experience, will quickly become a very real and a very valuable
efficiency, but if thwarted from outside will lead to friction,
insecurity and disorganisation.

It is particularly by their contempt, which they are at no pains to
disguise, for the competence that comes from contract and later from
habit, by their refusal to recognise the position of the head of the
family, that women every day and in every minute particular are training
their children to despise their father. Democracy seems bent on bringing
up its children to despise their parents. No other construction can be
put upon the facts, however good and innocent the motives. Just sum up
the facts. In the first place democracy denies that the living can be
guided by the dead; it is one of its fundamental axioms that no
generation should be tied and bound by its predecessor. What inference
can children be expected to draw from this except that they owe no
obedience to their father and mother?

Children have naturally only too great a tendency to look down on their
parents. They are proud of their physical superiority; they know that
their star is rising while that of their parents is setting. They are
imbued with the universal prejudice of modern humanity that _progress is
constant_ and that therefore whatever is of yesterday is _ex hypothesi_
inferior to that which is of to-day. They are driven also, as I am
constrained to believe, by a sort of Nemesis inspired by fear lest human
science and power should hurry forward too fast if the children were
content to pick up the burden of life where their parents left it, and
simply followed their fathers and did not insist on effacing all that
their fathers had done and beginning again--with the result that the
edifice never rises far above its foundations, and that children for
this and other reasons have a natural inclination to treat their parents
as Cassandras. Then, as it were to clench the argument, democracy is
ready with its teaching that each generation is independent of the
other, and that the dead have no lesson to impart to the living.

In the second place, democracy, applying the principle still further and
proclaiming the doctrine that the State is master of all, withdraws the
child from the family, as often and as completely as it can.
"Democracy," said Socrates, in one of his humorous dialogues, "is a
mountebank, a kidnapper of children. It snatches the child from its
family while he is playing, takes him far away, allows him no more to
see his family, teaches him many strange languages, drills him till his
joints are supple, paints his face and dresses him in ridiculous
clothes, and imparts to him all the mysteries of the acrobat's trade
until he is sufficiently dexterous to appear in public and amuse the
company by his tricks."

At all events democracy is determined to take the child away from his
family, to give him the education which it has chosen and not that which
the parents have chosen, and to teach him that he must not believe what
his parents teach him. It denies the competence of parents to rear their
children and puts forward its own competence, asserting that it is only
its own that has any value.

This is one of the principal causes of the divisions between fathers
and children in a democracy.

You may retort that democracy does not always succeed in its efforts to
separate children from their parents, because there is nothing to
prevent the children extending the contempt, which for such excellent
reasons they have been taught to entertain for their parents, to their
State-appointed teachers.

This is a most pertinent observation, for the general maxims of
democracy are just as likely to make pupils despise their masters as to
make sons despise their fathers. The master, too, represents in the eyes
of his pupil that past which has no connection with the present and
which by the law of progress is very inferior to the present. This is
true; but the end of all is that between the school which counteracts
the influence of the parents and the home which counteracts the
influence of the school, the child becomes a personage who is never
educated at all. He is in like case with a child who in the family
itself receives lessons, and what is more important, example, from a
mother who is religious and from a father who is an atheist. He is not
educated, he has had no sort of education. The only real education,
that is to say, the only transmission to the children of the ideas of
their parents consists of an education at home which is reinforced by
the instruction of masters chosen by the parents in accordance with
their own views. This is precisely the form of education to which
democracy refuses to be reconciled.

               *       *       *       *       *

There is a still more cogent reason why old men are neither respected
nor honoured in a democracy. Here is yet another efficiency formally
denied and formally set aside. An interesting treatise might be written
on the rise and fall of old men. Civilization has not been kind to them.
In primitive times, as among savage races to-day, old men were kings.
Gerontocracy, that is, government by the aged, is the most ancient form
of government. It is easy to understand why this should be. In primitive
ages, all knowledge was experience and the old men possessed all the
historical, social and political experience of the State. They were held
in great honour and listened to with the profoundest respect and
veneration, in fact with an almost superstitious reverence. Nietzsche
was thinking of those days when he said: "Respect for the aged is the
symbol of aristocracy," and when he added: "Respect for the aged is
respect for tradition," he was thinking of the reason for this
assumption. That the dead should rule the living was accepted
instinctively, and it was their nearness to death which evoked honour
for the aged.

At a later stage the old man shared in the civil government with
monarchy, aristocracy or oligarchy, and retained an almost complete
control of judicial affairs. His moral and technical efficiency were
still appreciated. His moral efficiency to his contemporaries consisted
in the fact that his passions were deadened and his judgment as
disinterested as was humanly possible. Even his obstinacy is rather an
advantage than otherwise. He is not liable to whims and fancies and
sudden gusts of temper or to external influence. His technical
efficiency is considerable, because he has seen and remembered much and
his mind has unconsciously drawn up a reference book of cases. As
history repeats itself with very slight alterations, every fresh case
which arises is already well known to him; it does not take him by
surprise and he has a solution at hand which only requires very slight
modification.

All this, however, is very ancient history. That which undermined the
authority of old men was the book. Books contain all science, equity,
jurisprudence and history better, it must be confessed, than the
memories of old men. One fine day the young men said: "The old men were
our books; now that we have books we have no further need for old men."

This was a mistake; the knowledge which is accumulated in books can
never be anything but the handmaiden of living science, the science
which is being constantly remodelled and corrected by living thought. A
book is a wise man paralysed; the wise man is a book which still thinks
and writes.

These ideas did not hold; the book superseded the old man, and the old
man no longer was a library to the nation.

Later still, for various reasons, the old men drifted from a position of
respect to one of ridicule. Undoubtedly they lend themselves to this;
they are obstinate, foolish, prosy, boring, crotchety and unpleasant to
look upon. Comic writers poked fun at these failings which are only too
self-evident and showered ridicule upon them. Then as the majority of
audiences is composed of young men, first of all because there are more
young men than old, and secondly because old men do not often go to the
theatre, authors of comic plays were certain of raising a laugh by
turning old men into ridicule, or rather by exposing only their
ridiculous characteristics.

At Athens and at Rome and probably elsewhere, the old man was one of the
principal grotesque characters. These things, as Rousseau pointed out,
have a great effect upon morals. Once the old man became a recognised
traditional stage-butt, his social authority had come to an end. In the
_de Senectute_ it is obvious that Cicero is running counter to the
stream in seeking to restore to favour a character about whom the public
is indifferent and for whom all he can do is to plead extenuating
circumstances.

It is a remarkable fact that even in mediæval epics, Charlemagne
himself, the emperor of the flowing beard, often plays a comic part. The
epic is invaded by the atmosphere of the fable.

During the Renaissance, the seventeenth and eighteenth centuries, the
old man is generally, though not invariably, held up to ridicule.

Molière takes his lead from Aristophanes and Plautus rather than from
Terence and is the scourge of old age as well as "the scourge of the
ridiculous"; he pursues the old as a hound his prey and never leaves
them in peace either in his poetry or his prose.

We must do this much justice to Rousseau that both he and his child, the
Revolution, tried to restore the old man to his former glory; he makes
honourable mention of him in his writings, and she gives him important
posts in public ceremonies and national fêtes. Therein were received the
ancient memories of Lacedæmon and of early Rome, combined with a form of
reaction against the days of Louis XIV and Louis XV.

But with the triumph of democracy the old man was finally banished to
the limbo of discredited things. Montesquieu's advice was quite
forgotten (see the context Laws, v, 8). He said that _in a democracy_
"nothing kept the standard of morals so high as that young men should
venerate the old. Both profit by it, the young because they respect the
old, and the old because they are confirmed in their respect for
themselves" (for the respect of the young is an assistance to the
self-respect of the aged).

Democracy has forgotten this advice, because it no longer believes in
tradition and believes too much in progress. Old men are the natural
upholders of tradition, and we must confess that an enthusiastic faith
in the value of what we call progress is not commonly their failing. For
this very reason their influence would be a most wholesome corrective to
the system, or rather to the attitude of mind, which despises the past
and sees in every change a step in the path of progress. But democracy
will not allow that it needs a corrective, and the old man, to it, is
only an enemy. The old man upholds tradition and has no enthusiasm for
progress, but beyond this he appeals for respect, first for himself,
then for religion, for glory, for his country and for the history of his
nation. Democracy is indifferent to the sentiment of respect, or rather
it lives in constant fear that the sentiment may be applied elsewhere.

Then what does democracy want for itself?

Not respect, but adoration, passion, devotion. We all like to see our
own sentiments as to ourselves repeated in the minds of others. The
crowd never respects, it loves, it yields to passion, enthusiasm,
fanaticism. It never respects even that which it loves.

It is quite natural that the masses should not care for old men. The
masses are young. How aptly does Horace's description of the young man
apply to the people!


  _Imberbis juvenis, tandem custode remoto
  Gaudet equis, canibusque et aprici gramine campi;
  Cereus in vitium flecti, monitoribus asper,
  Utilium tardus provisor, prodigus æris,
  Sublimis, cupidusque et amata relinquere pernix._


"Once free from the control of his tutors, the young man thinks of
nothing but horses, dogs and the Campus Martius, impressionable as wax
to every temptation, impatient of correction, unthrifty, extravagant,
presumptuous and light of love."

At all events respect has no meaning for the crowd, and when it rules,
we cannot from its example learn the lessons of respect. Democracy has
no love for the old; and it is interesting to note that the word
gerontocracy to which the ancients attached the most honourable meaning
is now only a term of ridicule, and is applied only to a government
which, because it is in the hands of old men, is therefore grotesque.

              *       *       *       *       *

This disappearance of respect, noted as we have seen by Plato, Aristotle
and Montesquieu as a morbid system, is, regard it how we will, a fact of
the gravest import. Kant has asked the question, what must we obey? What
criterion is there to tell us what to obey? What is there within us
which commands respect, which does not ask for love or fear, but for
respect alone? He has given us the answer. The feeling of respect is the
only thing that we can trust, and that will never fail us.

In society the only feelings we obey are those which win our respect,
and the men to whom we listen, and whom we honour, are those who inspire
respect. This is the only criterion which enables us to gauge correctly
the men and things to whom we owe, if not absolute obedience, at least
attention and deference. Old men are the nation's conscience, and it is
a conscience at times severe, morose, tiresome, obstinate,
over-scrupulous, dictatorial, and it repeats for ever the same old saws;
in other words a conscience; but conscience it is.

The comparison might be carried further with results that would be
advantageous as well as curious. We degrade and finally vitiate our
conscience if we do not respect its behests. Conscience then itself
becomes small and timid and humble, shamefaced, and at length a mere
whisper. Absolutely silent it can never be made.

It becomes sophisticated, it begins to employ the language of passion,
not of the vilest passions of our nature, but still the voice of
passion; it ceases to use the categoric imperative and tries to be
persuasive. It no longer raises the finger of command, but it seeks to
cajole with caressing hand.

Then it falls still lower, it affects indifference and scepticism and it
puts on the air of the trifler in order to insinuate a word of wisdom
into the seductive talk that is heard around it, and it holds language
somewhat as follows: "Probably everything has its good points and there
is something to be said for both vice and virtue, crime and honesty, sin
and innocence, rudeness and politeness, licence and purity. These are
all simply different forms of an activity which cannot be wholly wrong
in any of its manifestations; and it is precisely because every one of
these has its value that there may be nothing to lose in being honest,
nay, perhaps something to gain."

Nevertheless, a nation that does not respect its old men changes their
nature and despoils them of their beauty and integrity. How true is
Montesquieu's saying that the respect paid them by the young helps old
men to respect themselves! Old men who are not respected take no
interest in their natural duties; they cease to advise, or else they
only venture to advise indirectly, as though they were apologising for
their wisdom, or they affect a laxity of morals to enable them to
insinuate a surreptitious dose of worldly wisdom;--and worst of all in
view of the insignificant part assigned to them in society, old men will
nowadays decline to be old.




CHAPTER IX.

MANNERS.


If the worship of incompetence reverberates with a jarring note through
our domestic morals, it has an effect hardly less harmful on the social
relations of men in the wider theatre of public life. We often ask why
politeness is out of date, and everyone replies with a smile: "This is
democratic." So it is, but why should it be? Montesquieu remarks that
"to cast off the conventions of civility is to seek a method for putting
our faults at their ease." He adds the rather subtle distinction that
"politeness flatters the vices of others, and civility prevents us from
displaying our own. It is a barrier raised by men to prevent them from
corrupting each other." That which flatters vice can hardly be called
politeness, but is rather adulation. Civility and politeness are only
slightly different in degree; civility is cold and very respectful,
politeness has a suggestion of flattery. It graciously draws into
evidence the good qualities of our neighbour, not his failings, much
less his vices.

There is no doubt that civility and politeness are a delicate means of
showing respect to our fellow-men, and of communicating a wish to be
respected in turn. These things then are barriers, but barriers from
which we derive support, which separate and strengthen us, but which,
though holding us apart, do not keep us estranged from our neighbours.

It is also very true that if we release ourselves from these rules,
whether they are civility or politeness, we set our faults at liberty.
The basis of civility and politeness is respect for others and respect
for ourselves. As Abbé Barthélemy has very justly remarked: "In the
first class of citizens is to be found a spirit of decorum which makes
it evident that men respect themselves, and a spirit of politeness which
makes it evident that they also respect others." This is what Pascal
meant by saying that respect is our own inconvenience, and he explains
it thus, that to stand when our neighbour is seated, to remove our hat
when he is covered, though trifling acts of courtesy, are tokens of the
efforts we would willingly make on his behalf if an opportunity of being
really serviceable to him presented itself.

Politeness is a mark of respect and a promise of devotion.

All this is anti-democratic, because democracy does not recognise any
superiority, and therefore has no sympathy with respect and personal
devotion. Respect to others involves a recognition from us that we are
of less importance than they, and politeness to an equal requires from
us a courteous affectation that we consider him as our superior. This is
entirely contrary to the democratic ideal, which asserts that there is
no superiority anywhere. As for pretending to treat your equal as though
he were your superior, that involves a double hypocrisy, because it
requires a reciprocal hypocrisy on the part of your neighbour. You
praise his wit, only in order that he may return the compliment.

Without, however, insisting on this point, democracy will argue that
politeness is to be deprecated, because it not only recognises but
actually creates superiority. It treats an equal as a superior, as
though there were not enough discrepancies already without inventing
any more. It seems to imply that if inequality did not exist, it would
be necessary to invent it. It is tantamount to proclaiming that there
cannot be too much aristocracy. That is an opinion which democracy
cannot endure.

Considered as a promise of future devotion, politeness is equally
anti-democratic. The citizen owes no devotion to any person, he owes it
only to the community. It is no small matter to style yourself "your
most humble servant"; it means that you single out one man from among
many others and promise to serve him; it means that you acknowledge in
him some natural or social superiority, and according to democracy there
are no superiorities, social or natural, and if there were such a thing
as natural superiority, nature has no business to allow it. This is
tantamount to proclaiming a form of vassalage--a thing which is not to
be tolerated.

As to the absence of politeness considered as "a means of giving free
play to one's feelings," we recognise that in one sense this also is
essentially democratic. The democrat is not proud of or pleased with his
faults; not at all; only _ex hypothesi_ he does not believe in their
existence. A failing is an inferiority of one man in relation to
another; the word itself implies it; it means that something is lacking,
that one man has a thing which another has not. But all men are equal,
therefore, argues the democrat, I have no failing; therefore I need not
try to conceal and control my alleged failings, as they are at worst
merely mannerisms, and are possibly virtues.

The democrat, in fact, like young men, like most women, and like all
human beings who have begun to think but do not think very profoundly,
knows his failings and assumes that they are virtues. This is very
natural, for our faults are the most conspicuous parts of our character,
and when we are still at the self-satisfied stage it is our faults that
we cherish and admire. Consequently, politeness, in that it consists in
concealing our faults, is intolerable to a man who is impatient to
display qualities that to him appear commendable and worthy. The usual
reason why we do not correct our faults is that we mistake them for
qualities, and think that any practice which requires their concealment
must be quite absurdly tyrannical.

The democrat is therefore profoundly convinced of two things; first,
that all men are equal and that there is no such thing as inferiority or
failing, and secondly, that what men call faults are really natural
characteristics of great interest. He believes that faults are popular
prejudices invented by intriguers, priests, nobles and rulers, for their
own base purposes to inspire the poor with humility. He looks upon this
sense of inferiority as a curb on the people's power, all the more
potent that it works from within and has a paralysing effect on its
energy. He is persuaded that, from this point of view, politeness is an
aristocratic instrument of tyranny.

This explains why, when the wave of democracy swept over France, it
brought with it a perfect frenzy of rudeness, all the more curious in a
nation remarkable for courtesy. It was an affirmation that, appearances
notwithstanding, neither superiorities nor excellences of human
character had any real existence.

Rudeness is democratic.




CHAPTER X.

PROFESSIONAL CUSTOMS.


The contempt for efficiency is carried far even in the liberal
professions and in professional customs. We all know the story, perhaps
a mythical one, of the judge who said to an earnest young barrister who
was conscientiously elaborating a question of law: "Now, Mr. So and So,
we are not here to discuss questions of law but to settle this
business." He did not say this by way of jest; he wished to say: "The
courts no longer deliver judgment on the merits of a case according to
law, but according to equity and common sense. The intricacies of the
law are left to professors, so please when conducting a case do not
behave like a professor of law." This theory, which even in this mild
form would have horrified the ancients, is very prevalent nowadays in
legal circles. It has crept in as an infiltration, as one might call it,
from the democratic system.

A magistrate, nowadays, whatever remnant of the ancient feeling of caste
he may have retained, certainly does not consider himself bound by the
letter of the law, or by jurisprudence, the written tradition; when he
is anything more than a subordinate with no other idea of duty than
subservience to the Government, he is a democratic magistrate, a Heliast
of Athens; he delivers judgment according to the dictates of his
individual conscience; he does not consider himself as a member of a
learned body, bound to apply the decisions of that body, but as an
independent exponent of the truth.

An eccentric, but in truth very significant, example of the new attitude
of mind is to be found in the judge, who formally attributed to himself
the right to make law and who in his judgments made references, not to
existing laws, but to such vague generalities as appealed to him, or to
doctrines which he prophesied would _later on_ be embodied in the law.
His Code was the Code of the future.

The mere existence of such a man is of no particular importance, but the
fact that many people, even those partially enlightened, took him
seriously, that he was popular, and that a considerable faction thought
him a good judge, is most significant.

There is another much commoner sign of the times. The worst form of
incompetence is perhaps that which allows a man to be competent without
realising it, and, in criminal cases at least, this seems to be the
normal attitude of the majority of our magistrates.

We should read on this point a very curious pamphlet called _Le Pli
Professionnel_ (1909), by Marcel Lestranger, a provincial magistrate. It
is very pertinent to our subject. It shows plainly that the magistracy
nowadays, both the qualified stipendiaries and the bench of magistrates,
has lost all confidence in itself and is terrified of public opinion as
represented by newspapers, associations, political clubs and the man in
the street; the magistrate knows too, or thinks he knows, that promotion
depends, not on a reputation for severity as it used to do, but on a
reputation for indulgence.

He is confronted in the execution of his duty by forces which are always
in coalition against him; the public, almost always favourable to the
accused, the press, both local and Parisian, the so-called science of
judicial medicine, which is almost always disposed to consider the
accused as persons not responsible for their actions. He lives, too, in
constant terror of being mixed up in a miscarriage of justice, for
miscarriage of justice is now a sort of craze, and with a considerable
section of the public every conviction is a miscarriage of justice. And
so the magistrate of first instance never dares to sum up severely, and
the stipendiary never dares press his interrogations with firmness.

There are exceptions of course; but these exceptions, by the
astonishment which they excite, and by the reaction to which they give
rise, show sufficiently, indeed conclusively, that they are abnormal,
outside the new order of things, outside the new habits of the people.

More often than not the subordinate magistrate, whose business it is to
commit the prisoner for trial, acts with timidity and reserve,
apologetically attenuating the crime; he leaves loopholes of escape,
appeals in audible asides for indulgence, dwells on the uncertainty of
evidence. He demands indeed the prisoner's head but lives in terror lest
he obtain it.

The fact is what both he and the stipendiary desire is that the affair
should be settled by an acquittal, for an affair settled by an
acquittal is an affair buried. Stone-dead has no fellow; it is consigned
to oblivion. It can never be made the sort of affair which someone is
sure to declare is a miscarriage of justice, or which someone, animated
by private and political spite or merely for the sake of a jest, can
make into a ghost to haunt for ten or even fifteen years the unfortunate
magistrate who had to deal with it.

M. Lestranger tells a story which, from all the information I can glean
and from what I can remember hearing at the time, is absolutely true and
a perfect illustration of thousands of similar cases.

A poacher, aged nineteen, first outraged and then strangled in the woods
a peasant woman, the mother of a family. On this occasion there could be
no question of a miscarriage of justice or even of any suggestion of
such a thing, because the prisoner pleaded guilty. That is a great
point. In France every conviction that is not based upon the prisoner's
confession is a miscarriage of justice; but when the prisoner pleads
guilty there can be no incriminations of this sort, although there might
be, for false confessions are not unknown, but nothing of the sort is
ever put forward, and the case seemed to be quite straightforward.

But the magistrates were terrified that the prisoner would be condemned
to death. The crime was horrible, particularly in the eyes of a village
jury, whose wives and daughters were often obliged to work some distance
from the village. Moreover, there was a tiresome man, the widower of the
victim, thirsting for vengeance, who sang the praises of his wife and
brought his weeping son into court while he gave his evidence. The
president and the public prosecutor were in despair.

"I have done all I can," said the president to the public prosecutor. "I
have made the most of his youth. I have repeated 'only nineteen years of
age.' I have indeed done all I can."

"I have done all I can," said the public prosecutor to the president. "I
have not said a word about the punishment. I merely accused. I could not
plead for the defence. I have done my best."

At the close of the hearing the chief constable was very reassuring to
these gentlemen. "He is under twenty and he looked so respectable at
the enquiry. It is quite impossible that he should be condemned to death
in this quiet village. You will see, he will not be sentenced to capital
punishment."

He was not. The jury brought in a verdict of guilty with extenuating
circumstances. The magistrates recovered their tranquillity.

M. Lestranger's facts are supported by figures. Those who commit crimes
which excite pity, such as infanticide and abortion, are less and less
likely to be prosecuted, and if they are, they are frequently let off,
however flagrant the offence. The average number of acquittals during
the last twelve years is twenty-six per cent. A magistrate nowadays is a
St. Francis of Assize.

Either the magistrate does not believe in his own efficiency, or he
sacrifices it to his peace of mind, and he cares more for his own peace
of mind than for the public safety. The magistracy will soon be no more
than a _façade_, still imposing but not at all alarming.

There is already a very serious symptom of how little confidence the
crowd has in the wholesome severities of justice; the criminal caught in
the act is often lynched or almost lynched, because it is well known
that if he is not punished immediately, he is very likely to escape
punishment altogether.

--Yet this same crowd, in the form of a jury, is often, almost always,
very indulgent.--True, and that is because between the crime and the
assizes there is often an interval of six months. At the date of the
crime it is the misfortune of the victim that excites the crowd, at the
date of the assize it is the misfortune of the accused. Be this as it
may, the practice of lynching amounts to a formal accusation that both
magistrates and juries are over indulgent.

                *       *       *       *       *

The clergy even, who are more tenacious of tradition than any other
order in the State, are gradually becoming democratic to this extent,
that though by profession teachers of dogmas and mysteries, they now
teach only morality. In this way they try to get into closer touch with
the poor, and so have a greater hold upon them. Evidently they are not
altogether to blame. Only, when they cease to teach dogma and interpret
mysteries, they cease to be a learned body or to have the prestige of a
learned body. On the other hand they sink to the level of any other
philosophy, which teaches and explains morality, and illustrates it by
sacred examples just as well as any priesthood. The result is that the
people say to themselves "What need have we of priests? Moral
philosophers are good enough for us."

This Americanism is not very dangerous, in fact it does not matter, in
America, where there are very few lay moral philosophers; but it is a
very great danger in France, Italy and Belgium where their name is
legion.

                *       *       *       *       *

In every profession, to sum it all up, the root of the evil is this,
that we believe that mere dexterity and cunning are incomparably
superior to knowledge and that cleverness is infinitely more valuable
than sound learning. Those who follow professions believe this, and the
lay public that employs the professions is not dismayed by this attitude
of the professional class; and so things tend to that equality of
charlatanry to which democracy instinctively tends. Democracy does not
respect efficiency, but it soon will have no opportunity to respect it;
for efficiency is being destroyed and before long will have disappeared
altogether. There will soon be no difference between the judge and the
suitor, between the layman and the priest, the sick man and the
physician. The contempt which is felt for efficiency destroys it little
by little, and efficiency, accepting the situation, outruns the contempt
that is felt for it. The end will be that we shall all be only too much
of one opinion.




CHAPTER XI.

ATTEMPTED REMEDIES.


We have sought very conscientiously, and democrats themselves have
sought very conscientiously, to find remedies for this constitutional
disease of democracy. We have preserved certain bodies, relatively
aristocratic, as refuges, we would fain believe, of efficiency. We have
preserved for instance a Senate, elected by universal suffrage, not
directly, but in the second degree. We have preserved also a Parliament
(a Senate and a Chamber of Deputies), a floating aristocracy which is
continually being renewed. This is, however, in a sense an aristocracy
inasmuch as it stands between us and a direct and immediate government
of the people by the people.

These remedies are by no means to be despised, but we recognise that
they are very feeble, for the reason that democracy always eludes them.
By the care it takes to exclude efficiency, it has made the Chamber of
Deputies (with some few exceptions) a body resembling itself with
absolute fidelity both in respect of the superficial character of its
knowledge and the violence of its prejudices; with the result in my
opinion that the crowd might just as well govern directly and, without
the intervention of representatives, by means of the plebiscite.

The same thing applies to the Senate, though perhaps in a more direct
fashion. The Senate is chosen by the delegates of universal suffrage.
These delegates, however, are not chosen by a general universal suffrage
where each department would choose four or five hundred delegates, but
by the town councillors of each commune or parish. In these communes,
especially in the rural communes, the municipal councillors who are by
far the most numerous and, with regard to elections, the most
influential, are more or less completely dependent on the _préfets_. The
result is that the Senate is, practically, chosen by the _préfets_, that
is, by the Government, as used to be the case under the First and Second
Empire. The maker of the constitution made this arrangement for the
benefit of his own party, for he upheld authority; and he wanted the
Central Government to control the elections of the Senate. It has not
turned out as he intended. _Vos non vobis_, others have profited by his
device, as the following considerations will show.

It is well known that in France a deputy belonging to the opposition,
though sure of his constituents, and certain to be re-elected
indefinitely, who for private reasons wishes to be a senator, is obliged
to be civil to the Government in power, to abate his opposition, and to
make himself pleasant, if he wishes to avoid failure in his new
ambition. It is very inconvenient to have a strong and active opposition
in the Senate.

It comes back again to this, that we have a Senate not far removed from
one elected by universal suffrage.

Universal suffrage elects the Chamber of Deputies, the Chamber elects
the Government, and the Government elects the Senate. The Senate is
therefore an extremely feeble anti-democratic remedy, and if it were
intended as a check on democracy, it has not been a striking success.

If we really wish to have an upper chamber as competent as possible,
independent of the central authority, and relatively independent of
universal suffrage, we must establish a chamber elected by the great
constituent bodies of the nation, and also in my opinion, by universal
suffrage, but with modifications somewhat as follows. The whole nation,
divided for practical purposes into five or six large districts, should
elect five or six thousand delegates who in turn should elect three
hundred senators. There would then be no pressure from Government nor
any manufacture by the crowd of a representation fashioned in its own
image, and we should have a really select body composed of as much
competence as could be got in the country.

It is, however, exactly the opposite of this that is done, and the
French Senate is an extremely feeble, anti-democratic remedy.

It represents the rural democracy, arbitrarily guided and governed by
the democratic Government.

               *       *       *       *       *

Another remedy which has been given an equally conscientious trial is
the system of competitive examination, which is supposed to be a
guarantee for the ability of those who seek admission into government
service. The object of these examinations, which are extremely detailed
and complicated, is to test the ability of the candidate in every
particular, to give employment to merit and to exclude favouritism.

--You call that an anti-democratic remedy! It is as democratic as well
can be!--

Nay, pardon! It would be anti-monarchical if we lived under a monarchy,
anti-aristocratic if we lived under an aristocracy, and it is
anti-democratic because our lot is cast in a democracy. Competition for
public offices is a sort of co-optation. In fact it is co-optation pure
and simple. When I suggested that the magistracy should be chosen by the
magistrates, that is, the _Cour de Cassation_ by the magistrates and the
magistrates in turn by the _Cour de Cassation_, I was of course accused
of being paradoxical, as is always the case, when one suggests
something contrary to the usual custom. I was, however, only carrying a
little further the principle which is already applied to officials. In a
certain sense and to a large extent officials recruit their numbers by
co-optation.

It is true, they do not actually choose the officials, but they
eliminate the candidates whom they do not wish to have. Examination is
ostracism of the inefficient. The Government, of course, has to decide
who may be candidates, but its selection for employment is limited to
those of whom other officials (the officials who conduct the
examination) can approve. It is in fact co-optation.

The committee of examiners which admits a candidate to St. Cyr appoints
an officer. The committee which admits a candidate to the _École
Polytechnique_ appoints an officer or an engineer. A committee also
which refuses a candidate at either of these places is encroaching on
the National Sovereignty, because it is forbidding the National
Sovereignty to make of this young man an officer or an engineer. This is
co-optation. This is a guarantee of efficiency. Here a wall is raised
against incompetence, and against the jobbery under which incompetence
would profit.

It is hardly necessary for me to add that this co-optation is limited to
a very narrow field of operation. It is confined in fact to the
threshold of a man's career. Once the candidate has been consecrated
official, by a board of examining officials, he belongs, both as regards
advancement, promotion and the reverse, to the central authority alone,
except in certain cases. The co-optation of officials is merely a
co-optation by elimination. The elimination is made once and for all,
and the non-eliminated (_i.e._, the successful candidate) steps at once
into the toils of the Government, that is, into the toils of popular
electioneering and party politics, when all the abuses which I have
enumerated can and do arise. To be fair I had of course to point out
that we had tried to invent some slight barriers against the omnipotence
of incompetence, which prevent it being absolutely supreme.

Unfortunately these prophylactic measures are very badly organised, and,
far from being capable of amendment, ought to be completely
revolutionised.

The examination system in our country is founded on a misconception, I
mean on the confusion between knowledge and competence. We search
conscientiously for competence or efficiency, and we believe that we
have found it when we find knowledge, but that is an error. An
examination requires from a candidate that he shall know, and
competition demands that he shall know more than the others, but
that is almost all that examination and competition require of him.
Therefrom results one of the most painful open sores of our
civilisation,--preparation for examinations.

Preparation for examination is responsible for intellectual indigestion,
for minds overloaded with useless information, and for a system of
cramming, which at once takes the heart out of men, perhaps with good
ability, just at the age when their mental activity is most keen; which,
further, as the result of this surfeit, disgusts for the rest of his
life and renders impotent for all intellectual effort, the unfortunate
patient who has been condemned to undergo this treatment for five,
eight, and sometimes ten years of his youth.

I am satisfied, if I may be allowed to speak of myself in order to
support my argument by an instance well known to me, that, if I have
been able to work from the age of twenty-five to that of sixty-three, it
is because I have never succeeded except very moderately, and I am proud
of it, in competitive examinations. Being of a curious turn of mind I
have been interested in the subject set in the syllabus, but in other
matters also, and the syllabus has been neglected. I sometimes passed,
more often I failed, with the result that at twenty-six I was behind my
contemporaries, but I was not overworked, broken down, and utterly sick
of all intellectual effort. I admit that some of my contemporaries who
never failed in an examination, and who passed them all with great
brilliance, have worked as hard as I have up to sixty, but they are
extremely few.

The curious thing is that the results, not perhaps disastrous, but
obviously very unsatisfactory, of this examination system do not lead us
to abandon it (that perhaps would be an extreme measure), but make us
aggravate and complicate it. Legal and medical examinations are much
"stiffer" than they used to be, and they require a greater physical
effort, but without requiring or obtaining any greater intellectual
value. In truth, one might say, examination is nothing more than a test
of good health, and it is a very searching test, for it often succeeds
in destroying it.

Here is an example which I know well. It is necessary, if a man desire
to gain distinction as a professor of secondary education, that he
should be a bachelor, a licentiate, an _agrégé_ or a doctor. This is a
qualification that counts, and it means ten examinations or
competitions, two for the first half of the bachelor's degree, two for
the second, two for the licentiate, two for _agrégé_, two for the
doctor's degree. This, moreover, does not appear to be enough. Between
the second part of the bachelor's degree and the licentiate's degree
there is normally an interval of two years; between the licentiate and
the _agrégation_ two years, and between the _agrégation_ and the
doctor's degree there is generally three or four years. You perceive the
danger! Between the _licence_ and the _agrégation_, to go no further at
present, the future professor has two whole years to himself. That is to
say, that during the first of these two years he will work alone. He can
work freely, he can study in what direction he pleases, without thinking
of an examination at the end of twelve months; he has escaped for the
moment from the servitude of the syllabus. The prospect makes us shudder
with apprehension. It is sadly to be feared that the young man may take
a rest and draw breath, or worse still he may be carried into some
extraneous study by his personal aptitudes or tastes. The personality of
the candidate has here an opening, a moment at which it has a
possibility of asserting itself. That must be stopped at all costs.

The authorities, therefore, have put in an intermediate examination
between the _licence_ and the _agrégation_. The examination, it is true,
is on a subject chosen by the candidate himself; so much it is only fair
to admit. The subject chosen, however, must be submitted to the
professors. Their advice and indeed assistance must be invited. The
result, if not the object, of this examination is to prevent the
candidate, during this perilous year of liberty, from developing
original ideas of his own and acting on them.

_One examination every year for ten years_--that is the ideal of the
modern professor for the future professors who are in course of being
trained. Between the second part of the bachelor's degree and the
licentiate, as there is there an interval of two years, they will
presently perceive that there ought to be an examination at the end of
the first year, and we shall have certificates of study in
intermediate, secondary, higher subjects. Between the _agrégation_ and
the _doctorat_, there are four years, and naturally we shall want three
examinations just to see how the future professor is getting on with his
theses, to encumber him with assistance and to prevent him doing them
alone; first examination called the _Bibliography of the Theses for the
Doctorat_, second examination called the _Methodology of the Doctorat_,
third examination called the _Preparation for the Sustaining of the
Thesis_, and then the examination for the doctor's degree itself.

In this way the desired object is attained. Between the ages of
seventeen and twenty-seven or thirty the examinee will have had to
undergo sixteen examinations. He will never have worked alone. He will
always have worked, for periods of twelve months, on a syllabus, for an
examination, with a view of pleasing such and such professors, modelling
himself on their views, their conceptions, their general ideas, their
eccentricities, aided by them, influenced by them, never knowing, and
feeling he ought not to know, not wishing to know, and running a great
risk if he did know, and forming habits for his whole life so that he
may never know what he thinks himself, what he imagines himself, what he
seeks and would like to seek of his own motion, or what he ought himself
to try to be. He will take up all this after he is thirty.

Not a vestige of personality or original thought till the moment when it
is too late for it to appear, that is the maxim!

Whence comes this frenzy, this _examino mania_? When one comes to think
of it, it seems to be a simple case of _Dandino-mania_. Dandin says with
great determination "I mean to go and judge." The professor of a certain
age means to go and examine. He no longer loves to profess, he loves to
be always examining. This is very natural. Professing, he is judged;
examining, he judges. The one is always much pleasanter than the other.
For a professor, to sweat in harness, to feel oneself being examined,
that is, criticised, discussed, held up to judgment, and chaffed by an
audience of students and amateurs, ceases at a certain age to be
altogether pleasant; on the other hand to examine, to sit on the throne
with all the majesty of a judge, to have only to criticise and not to
produce, to intervene only when the victim stumbles, and to let him
know that he has made a slip, to hold the student for the whole year
under the salutary terror of an approaching examination, to remind him
that he may need help and must by no means displease his professor--all
this is very agreeable and makes up for many of the worries of the
teaching profession. The examination mania proceeds partly from the
terror of being oneself examined, and partly from the pleasure of
examining others.

All this is true, but there is more than this. The precocious
development of early talent and originality is the thing which strangely
terrifies these examination-maniacs. They have a horror of the man who
teaches himself. They have a horror of any one who ventures to think for
himself and to enquire for himself at twenty-five years of age. They
want, like an old hen, to mother the young mind as long as possible.
They will not let it find its own feet, till very late, and till, as the
scoffer might well say, its limbs are absolutely atrophied. I do not say
that they are wrong. The man who has taught himself is apt to be a vain,
conceited fellow who takes pleasure in thinking for himself, and has
an absolute delight in despising the thoughts of others. It is, however,
no less the fact, that it is among these self-taught men that we find
those vigorous spirits who venture boldly beyond the domain of human
science and extend its frontier. The question then is which is best, to
favour all these troublesome self-taught people in the hope of finding
some good ones among them, or by crossing and worrying them to run the
risk of destroying the good as well as the bad. I am myself strongly in
favour of the first of these alternatives. It is better to let all go
their own way, even though pretenders to originality come to grief, a
thing that matters very little. Minds that are truly original will
develop themselves and find room for the expansion of all their powers.

But here,--take note how the democratic spirit comes in everywhere--the
question of numbers is raised. Ten times more numerous, I am told, are
the pretenders to originality whom we save from themselves by discipline
than the true geniuses whose wings we clip.

I reply that, in matters intellectual, questions of figures do not
count. An original spirit strangled is a loss which is not compensated
by the rescue of ten fools from worse excesses of folly. An original
spirit left free to be himself is worth more than ten fools whose folly
is partially restrained.

Nietzsche has well said: "Modern education consists in smothering the
exceptional in favour of the normal. It consists in directing the mind
away from the exceptional into the channel of the average." This ought
not to be. I do not say that education should do the opposite of all
this. Oh no, far from that. It is not the business of education to look
for exceptional genius, or to help in its creation. Exceptional genius
is born of itself and it has no need of such assistance. But even less
is it the business of education to regard the exceptional with terror,
and to take every means possible, even the most barbarous and most
detailed, to prevent it as long as possible from coming to the light.

Education ought to draw all that it can out of mediocrity, and to
respect originality as much as it can. It ought never to attempt to turn
mediocrity into originality, nor to reduce originality to the level of
mediocrity.

And how can all this be done? By an intervention that is always
discreet, and sometimes by non-intervention.

At the present moment its policy is equally distant from
non-intervention and from an intervention that is discreet.

It is in this way that the very institution which we have invented to
safeguard efficiency contributes not a little to the triumph of its
opposite. These victims of examination are competent in respect of
knowledge, instruction and technical proficiency. They are incompetent
in respect of intellectual value, often, though perhaps not so often as
formerly, in respect of moral value.

As far as their intellectual value is concerned, they have very
frequently no mental initiative. It has been cramped, hidden away, and
trampled down. If it ever existed, it exists now no longer. They are all
their days merely instruments. They have been taught many things,
especially intellectual obedience. They continue to obey intellectually,
their brain acts like well made and well lubricated machinery. "The
difference between the novel and the play," said Brunetière, "is that in
the play the characters act, in the novel they are acted." I do not
know if this be true, but of the functionary we might say as often as
not, he does not think, he is thought.

The official also is incompetent, though less and less often, in respect
of moral worth. By the exercise of intellectual obedience, he has been
trained to moral obedience also and he is little disposed to assert his
independence. Observe how everything tends to this end. This method of
co-opting officials by means of elimination, as I have said, operates
only, as I have also shown, at the outset of the official's career. From
this moment onwards the functionary must depend on the Government only,
his whole preparation during ten years of education has been calculated
to ensure his absolute dependence on his official directors. So far
good, perhaps a little too good. It would have been well if the
education of the functionary had left him, together with a little
originality of mind, a little originality of character as well.

             *       *       *       *       *

We have sought, very conscientiously also, and, I may even say, with an
admirable enthusiasm, yet another remedy for the faults of democracy,
another remedy for its incompetence. It is said: "The crowd is
incompetent, so be it, it is necessary to enlighten it. Primary
education, spread broadcast, is the solution of every difficulty, and
provides an answer to every question."

From this argument aristocrats have derived some little amusement. "How
is this?" they exclaimed, "what is the meaning of this paradox? You are
democrats and that means that you attribute political excellence,
'political virtue,' as we used to say, to the crowd, that is to
ignorance. Why then do you wish to enlighten the crowd, that is to
destroy the very virtue which, on your own showing, is the cause of its
superiority?" The democrats reply that the crowd, even as it is, is
already very preferable to aristocracy, and that it will be still more
so when it has received instruction. They resolve the apparent
contradiction by the argument _a fortiori_.

At all events, the democrats set to work most vigorously on the
education of the people. The result is that the people is much better
educated than formerly, and I am one of those who regard this result as
excellent; but the further result is, that the people is saturated
with false ideas, and this is less comforting.

Ancient republics had their demagogues, their orators, who inflamed the
evil qualities of the people, by bestowing on them high-sounding names
and by flattery. The great democracy of modern times has its demagogues.
These are its elementary school teachers. They come of the people, are
proud to belong to it, for which of course no one can blame them, they
distrust everything that is not the people, they are all the more of the
people because among the people they are intellectually in the first
rank while elsewhere they are of secondary importance; and what men love
is not the group of which they form a part, but the group of which they
are the chief. They are, therefore, profoundly democratic.

So far nothing could be better. But it is a narrow form of democratic
sentiment which they hold, for they are only half-educated, or rather
(for who is completely educated or even well educated?), because they
have only received a rudimentary education. Rudimentary education may
perhaps make us capable of having one idea, it certainly renders us
incapable of having two. The man of rudimentary education is always the
man of one single idea and of one fixed idea. He has few doubts. Now the
wise man doubts often, the ignorant man seldom, the fool never. The man
of one idea is more or less impermeable to any process of reasoning that
is foreign to this idea. An Indian author has said: "You can convince
the wise; you can convince, with more difficulty, the ignorant; the
half-educated, never."

Now no one ever convinces the elementary schoolmaster. He is confirmed
in his convictions by defending, and still more by discussing them. He
is the slave of his opinion. He does not possess it always quite
clearly, but it possesses him. He loves it with all his soul, as a
priest his religion, because it is the truth, because it is beautiful,
because it has been persecuted, and because it means the salvation of
the world. He would enjoy its triumph but he yearns still more to be a
martyr in its cause.

He is a convinced democrat and a sentimental democrat. His conviction
forms a solid basis for his sentiment, and his sentiment kindles to a
white heat his conviction. His conviction makes him turn a deaf ear to
every objection, his sentiment inspires him with hatred for his
adversary. For him the man who is not a democrat is wrong, and further,
to him an object of hatred. In his eyes the distance between himself and
the aristocrat is as the distance between truth and error, nay between
good and evil, between honour and dishonour. The schoolmaster is the
fanatic vassal of democracy.

Then, as he is a man of one idea, he is single-minded, narrowly logical,
and logical to the utmost extreme. He goes straight forward where his
argument leads. An idea which admits neither qualification nor question
can go far in a very short space of time. And the schoolmaster drives
all his democratic principles to their natural and logical conclusion.

He develops these principles and all that they imply by the sheer force
of what he calls his "reasoning reason," and it appears to him to be not
only natural but salutary to seek their realisation. Everything of which
the principle is good is good itself, and no one but Montesquieu could
ever believe that an institution could be ruined by the excess of the
principle in which its merit consists.

The schoolmaster, therefore, deduces their logical consequences from the
two great democratic principles, the sovereignty of the nation, and
equality; he deduces them rigorously, and arrives at the following
conclusions.

The people alone is sovereign. Therefore, though there can be individual
liberty and liberty of association, there ought to be only such
individual liberty and liberty of association as the people permits.
Liberty cannot be and ought not to be anything more than a thing
tolerated by the sovereign people. The individual may think, speak,
write, and act as he pleases, but only so far as the people will allow
him; for if he can do these things with absolute freedom, or even with
limitations which are not imposed by the people, he becomes the
sovereign power, or the power which fixed the limits of his freedom
becomes the sovereign, and the sovereignty of the people disappears.

This brings us back to the simple definition that liberty is the right
to do what we please within the limits of the law. And who makes the
law? The people. Liberty is then the right to do everything which the
people permits us to do. Nothing more; if we attempt to go beyond this,
the sovereignty of the individual begins, and the sovereignty of the
people disappears.

--But to have liberty to do only what the people permits, this is to be
free as we were under Louis XIV.--and that is not to be free at all!

So be it. There will indeed be no liberty unless the law permit it.
Surely you do not wish to be free in opposition to the law?

--The law may be tyrannical. It is tyrannical if it is unjust.--

The law has the right to be unjust. Otherwise the sovereignty of the
people would be limited and this must not be.

--Fundamental and constitutional laws might be devised to limit this
sovereignty of the people in order to guarantee such and such of the
liberties for the individual.--

And the people would then be tied! The sovereignty of the people would
be suppressed! No, the people cannot be tied. The sovereignty of the
people is fundamental and must be left intact.

--Then there will be no individual liberty?--

Only such a measure as the people will tolerate.

--Then there will be no liberty of association?

Still less; for an association is in itself a limitation of the
sovereignty of the nation. It has its own laws, which from a democratic
point of view is an absurd and monstrous incongruity. The right of
association limits the national sovereignty, just as would a free town
or sanctuary of refuge. It limits the nation, and pulls it up short in
face of its closed doors. It is a State within a State; where there is
association, there arises at once a source of organisation other than
the great organism of the popular will. It is like an animal which lives
some sort of independent life within another animal larger than itself
and which, living on that other animal, is still independent of it. In
fact there can be only one association, the association of the nation,
otherwise the sovereignty of the nation is limited, that is, destroyed.
No liberty of association can then exist.

Associations of course will exist which the people will tolerate, but
their right of existence is always revocable and they are always liable
to be dissolved and destroyed. Otherwise the national sovereignty would
be held to abdicate and it can never abdicate.

--Ah! but there is one association, at least, which to some extent is
sacred, and which the sovereignty of the people is bound to respect. I
mean the family. The father is the head of the family, he educates his
children and brings them up as he thinks best, till they come to man's
estate.--

Nay, that will not pass! For here again we have a limitation of the
sovereignty of the nation. The child does not belong to his father. If
this were so, at the threshold of each home the sovereignty of the
people would be arrested, which means that it would cease to exist
anywhere. The child, like the man, belongs to the people. He belongs to
it, in the sense that he must not be a member of an association which
might dare to think differently from the people, or perhaps even harbour
ideas in contradiction to the thought of the people. It would indeed be
dangerous to leave our future citizens for twenty years outside the
national thought, which is the same thing as being outside the
community. Imagine five or six bees brought up apart, outside the laws,
regulations, and constitution of the hive; imagine further that of these
groups of bees there were several hundreds in the hive. The result would
be the destruction of the hive.

It is _above all things_ in the family that the sovereignty of the
people ought to prevail. It ought above all things to refuse to
recognise the association of the family, and to wage war against it
wherever it finds it. It should leave to parents the right of embracing
their children, but nothing more. The right to educate them in ideas
perhaps contrary to those of their parents belongs to the people, which,
here as well as elsewhere, perhaps even more than elsewhere for the
interests at stake are more important, must be absolutely sovereign.

This, then, is what the schoolmaster, with a relentless logic which
appears to me to be irresistible, deduces from the principle of the
national sovereignty.

From the principle of equality he deduces another point. "All men are
equal by nature and before the law." That is to say, if there were
justice, all men ought to have been equal by nature, and further, if
there is to be justice, all men ought to be equal before the law.

Very obviously, however, all men are not equal before the law, and they
are not equal by nature. Very well then, we must make them so.

They are not equal before the law. They appear to be so, but they are
not. The rich man, even supposing that the magistrates are perfectly and
strictly honest, by reason of the fact that he can remunerate the best
solicitors, advocates, and witnesses, by reason further of the fact that
he intimidates by his influence all those who could appear against him,
is not in every respect the equal of the poor man before the law.

Even less does this equality exist in the presence of that union of
constituted social forces which we call society. In this respect the
rich man will be the "influential man"; the "man well connected," the
man on whom no one depends, but whom no one likes to cross or to
contradict. There is, between the rich and the poor man, however equal
we may pretend them to be before the law, the difference between the man
who gives orders and the man who is obliged to obey. _Real_ equality,
in society, in presence of society and even in presence of the law,
only exists where there is neither rich nor poor.

But there will always be rich and poor, as long as the institution of
inheritance remains. Abolish inheritance therefore!

But, even with inheritance abolished, there will still be rich and poor.
The man who can make his fortune rapidly will be a strong man relatively
to the man who can not make a fortune, and, I would have you note it,
even when we have abolished inheritance, the son of the strong man,
during the life of his father, will be strong himself, so that even if
we abolish inheritance, a privilege, namely, the privilege of birth will
still exist and equality will not exist.

There is only one state of affairs under which equality is possible,
that is when no one possesses and no one can acquire anything. The only
social policy so devised that no one can possess and no one can acquire
anything is the policy of a community of goods, that is Communism or
Collectivism. Collectivism is nothing very wonderful. Collectivism is
equality; and equality is collectivism, otherwise our equality will be
nothing but a phantom and an hypocrisy. Every one who is a convinced and
sincere _egalitarian_, and who takes the trouble to think, is forced to
be a collectivist. Bonald asked very wittily: "Do you know what is a
deist? It is a man who has not lived long enough to be an atheist." We
in our turn ask: "Do you know what is an anti-collectivist democrat? It
is a man who has not lived long enough to be a collectivist, or who,
having lived long enough, has never taken the trouble to think, and to
perceive what are the necessary consequences of his own principles."

But surely collectivism is a chimæra, an utopia, a thing impossible.
Certainly it is impossible in the sense that in the country which adopts
it the source of all initiative will be destroyed. No man will make an
effort to improve his position, since it must never be improved. The
whole country will become one of those stagnant pools to which one of
our ministers lately referred. Everyone having become an official,
everyone will realise the ideal of the official which the Goncourts have
very neatly described. "The good official," they say, "is the man who
combines laziness with extreme accuracy." It is a definitive
definition. The country that reformed itself in this way would be
conquered at the end of ten years by some neighbouring people more or
less ambitious.

That admits of no question; but what does it prove? That collectivism is
only impossible because it is only possible if established in every
country at once. Very well, and in order to establish it in every
country at once, only one thing is needful, namely, that there shall no
longer be distinct and separate countries and no longer any
nationalities. It surely will not answer to establish collectivism
before the abolition of nationalities, since, once established, it will
serve no purpose except to bring into prominent relief the vast
superiority of countries which have not adopted collectivism. We must,
therefore, take our problems in order and abolish nationalities before
we can establish collectivism.

Now if nations organise themselves against nature (the nature that, the
schoolmaster assumes, makes all men equal), if instinctively they
organise themselves in a hierarchy which is aristocratic, if they have
their leaders and their subordinates, their stronger and their weaker
members, it is because this arrangement is necessary in a camp, and each
nation feels that it is a camp. If each feels that it is a camp, it is
simply because there are other nations round it, because it feels and
knows that there are others round it. When there are no longer other
nations, each nation will organise itself no longer against nature, but
naturally, that is to say on _egalitarian_ principles. Nature perhaps
strictly speaking is not _egalitarian_, but it tends towards equality in
the sense that it produces many more, indeed infinitely more,
mediocrities than superior intelligences.

Thus equality demands the abolition of inheritance, and the equality of
possessions. Equality of possessions necessitates collectivism, and
collectivism requires the abolition of nationalities. We are
_egalitarians_, then collectivists, and by logical consequence
anti-patriots.

So argue the great majority of school teachers, with an absolute logic,
in my opinion, irrefutable, with the logic which takes no account of
facts, and which only takes account of its own principle and of itself.
So they will all argue to-morrow, if they continue, as it is probable
they will continue, to be very excellent dialecticians.

Will they go back to the premises and say, that if the sovereignty of
the people and equality lead logically and imperatively to these
conclusions, it is perhaps because the sovereignty of the people and
equality are false ideas, and because these conclusions prove them to be
false? This is a course not likely to be taken, for the sovereignty of
the people and the principle of equality are something more than general
ideas, they are sentiments.

They are sentiments which have become ideas, as is the case doubtless
with all general ideas, and they are sentiments of great strength. The
sovereignty of the people is the truth for him who believes in it,
because it ought to be true, because it is a thing as full of majesty
for him as was Cæsar in all his pomp for the ancient Roman, or Louis
XIV. in all his glory for the man of the seventeenth century.

Equality is truth for him who believes, because it ought to be true,
because it is justice, and because it would be infamous if justice and
truth were not one. For the democrat, the world has ever been rising
gradually, since its creation, towards the sovereignty of the people and
the doctrine of equality; the latter contains the former, the former is
destined to found the latter and has this mission for its purpose in
life; together they constitute civilisation, and if they are not
attained, there is a relapse into barbarism.

They are dogmas of faith. A dogma is an overmastering sentiment which
has found expression in a formula. From these two dogmas everything that
can be deduced without breach of logic is truth which it is our right
and duty to proclaim.

We must add that the schoolmaster is urged in this same direction by
sentiments of a less general character, which nevertheless have an
influence of their own. He is placed in his commune in direct opposition
to the priest, the only person very often who is, like himself, in that
place a man of some little education. Hence rivalry and a struggle for
influence. Now the priest, by a series of historical incidents, is a
more or less warm partisan sometimes of monarchy but almost always of
aristocracy. He is a member of a body that once was an estate of the
realm, and he is persuaded that his corporation is still an estate of
the realm, notwithstanding all that has happened. If the existing order
is regulated by the _concordat_, the existing order recognises his
corporation as a body legitimatised by the State, since it treats it on
the same terms as the magistracy and the army. If the existing order is
one based on the separation of State and Church, his corporation appears
to him still more to be an estate of the realm, because being forced
into an attitude of solid organisation, and recognising no limitations
of frontier, it becomes a collective personage which, not without peril,
but also not without a certain measure of success, has often ventured to
cross swords with the State itself.

As the priest then belongs to an order endowed with an historic
authority which is nevertheless distinct from, and in no wise a
delegation from, the authority of the people, the priest cannot fail
more or less definitely and consciously to adopt an attitude of mind
favourable to aristocracy.

The school teacher, his rival, is thrown then all the more inevitably
towards the adoption of democratic principles, and he embraces them with
a fervour into which enters jealousy quite as much as conviction. They
mean more to him than even to an eighteenth century philosopher, because
he has a much greater personal interest in believing them, the interest
of personal dislike and animosity; for it is his belief that everything
taught by the priest is the pure invention of ingenious oppressors who
wish to enslave the people in order to consolidate their own tyranny;
and that is his reason for professing philosophical ideas resuscitated
from the teaching of Diderot, and Holbach. For the school teacher it is
almost inconceivable that the priest should be anything but a rascal.

"Atheism is aristocratic," said Robespierre, thinking of Rousseau.
Atheism is democratic, say our present-day school teachers. Whence comes
this difference of opinion? First because it was fashionable among the
great lords of the eighteenth century to be libertines and
free-thinkers, but among the people the belief in God was unanimous.
Secondly, because the priests of our day, for the reasons which I have
given and from remembrance of the persecutions suffered by their Church
at the date of the first triumphs of democracy, have remained
aristocrats or have become so even more firmly than they ever were
before. Atheism then has become democratic as a weapon against the
deists who are generally aristocrats.

Besides, atheism fits in very well, whatever Robespierre may have
thought, with the general sentiments of the baser demagogy. To be
restrained by nothing, to be limited by nothing, that is the dominant
idea of the people, or rather it is the dominant idea of the democrat
for the people, that it should be restrained by nothing and limited by
nothing in its sovereign power. Now God is a limit, God is a restraint.
And just as the democrat will not admit of a secular constitution which
the people could not destroy and which would prevent him from making bad
laws; just as the democrat will not submit--if we may adopt the
terminology of Aristotle--to being governed by _laws_, to be governed
that is by an ancient body of law which would check the people and
obstruct it in its daily fabrication of _decrees_; so just in the same
spirit the democrat does not admit of a God Who has issued His
commandments, Who has issued His body of laws, anterior and superior to
all the laws and all the decrees of men, and Who sets His limit on the
legislative eccentricities of the people, on its capricious omnipotence,
in a word, on the sovereignty of the people.

After Sedan, Bismarck was asked: "Now that Napoleon has fallen, on whom
do you make war?" He replied: "On Louis XIV." So the democrat questioned
on his atheism could reply: "I am warring against Moses."

This is the origin of the atheism of democrats and schoolmasters. This
is the origin of the formula: "Neither God nor Master," which for the
anarchist requires no correction nor supplement, which for the democrat
has only to be modified: "Neither God nor Master, save the People."

At the end of one of his great political speeches in 1849 or 1850,
Victor Hugo said: "In the future there will only be two powers; the
People and God." The modern democrat has persuaded himself that if there
be a God, the sovereignty of the people is infringed, if he believe in
Him.

Lastly, the school teacher is confirmed in his democratic sentiments, in
all his democratic sentiments, by the political position which has been
made for him in France. It is a strange thing, a disconcerting anomaly,
that the Governments of the nineteenth century (especially, we must do
it this justice, the present Government), have very handsomely respected
the liberty of professors of higher education, and of secondary
education, and have not in the very slightest degree respected the
liberty of the teachers of elementary education. The professor of higher
education, especially since 1870, can teach exactly what he pleases,
except immorality and contempt of our country and its laws. He can even
discuss our laws, provided always that he maintains the principle that,
such as they are, they ought to be obeyed till they are repealed. His
liberty as to his opinions political, social and religious is complete.
It is only occasionally constrained by the disorderly demonstrations of
his students. The professor of secondary education enjoys a liberty
almost equally wide. He is subject, but only in an extremely liberal
fashion, to a programme or syllabus of studies. As to the spirit in
which he conducts his work he is practically never molested. He is
given a free hand.

Nor has it ever occurred to any Government to ask a professor of higher
and secondary education how he votes at political elections, still less
to require him to canvass in favour of the candidates agreeable to the
Government.

When, however, we pass to elementary education we see everything is
changed. The elementary teacher is not appointed by his natural chief,
the _recteur_ or Minister of Public Education, he is appointed by the
_préfet_, that is by the Minister of the Interior, the political head of
the Government. In other words, this is the same process as the
appointment of officials by the people, described a few pages back, but
with one intermediary the less. It is pre-eminently the Minister of the
Interior who represents the political will of the nation at any given
date. And it is the Minister of the Interior who through his _préfets_
appoints the elementary school teacher. It is then the political will of
the nation which chooses the school teachers. It would be impossible to
convey to them more clearly (which is only fair, for people should be
made to understand their duties) that they are chosen for
considerations of politics and that they ought to consider themselves as
political agents.

And indeed they are nothing else, or perhaps we should say they are
something else but above all they are politicians. The schoolteachers
depend on the _préfets_ and the _préfets_ depend much on the deputies,
yet it is not the deputies who appoint them, but it is they who can
remove them, who can get them promoted or disgraced, who by constant
removals can reduce them to destitution. Surely, every candid person
will exclaim, given the difficult and scandalous situation in which they
are put by the hand which appoints them, they ought at least to have the
guarantee and assurance, very relative and ineffectual though it be, of
irremovability. But they have not got it. The professors of higher
education who do not require it have got it, the professors of secondary
education have it to all intents and purposes. The elementary school
teacher has it not.

He is, therefore, delivered over to the politicians who make of him an
electioneering agent, who reckon him as such, and who would never pardon
him if he failed them.

The result is that the majority of school teachers are demagogues
because they like it, and with magnificent enthusiasm and passion. The
minority who have no turn for demagogy are demagogues though they do not
like it, and because they are forced by necessity.

Even those who have no disposition that way become demagogues in the
end, for that is the way of the world. "In the heat of the _mêlée_,"
said Augier, "there are no mercenaries." Our school teachers, thrown,
sometimes against their will, into the battle, forced at least to appear
to be fighting, receive knocks and when they have received them, they
become attached to the cause on whose behalf they have suffered. We
always end by having the opinions which are attributed to us, and being
taken for a demagogue the moment he arrives at his village, the young
school teacher, not daring to say anything to the contrary, and being
very ill received by all other parties, naturally becomes a demagogue
with some show of conviction the very next year.

             *       *       *       *       *

So the democracy receives no instruction that does not confirm and
strengthen it in its errors.

For its good some one ought to teach it not to believe itself
omnipotent, to have scruples as to its omnipotence, and to believe that
this omnipotence should have defined limits; it is taught without
reserve the dogma of the unlimited sovereignty of the people.

For its good it should believe that equality is so contrary to nature
that we have no right to torture nature in order to establish real
equality among men, and that the people which has established such a
state of things, which is quite possible, must succumb to the fate of
those who try to live exactly in opposition to the laws of nature.
Instead, it is taught, and it is true enough, that equality is not
possible, if it is not complete, if it is not thorough, that it ought to
be applied to differences of fortune, social position, intelligence,
perhaps even to our stature and personal appearance, and that no effort
should be spared to bring all things to one absolute level.

For its good, since it is natural enough that it should dislike heavy
taxation, sentiments of patriotism should be reinforced; it is taught on
the contrary that military service is a painful legacy left by a hateful
and barbarous past, and that it ought to disappear very soon before the
warming rays of a peaceful civilisation.

In a word, to use again the language of Aristotle, the pure wine of
democracy is poured out to the people as it was by the demagogues to the
Athenians; and from the quarter whence a remedy might have been expected
there come only incitements to deeper intoxication.

Aristotle has made yet another wise and profound observation on the
question of equality: "_We must establish equality_," he said, "_in the
passions rather than in the fortunes of men._" And he adds: "And this
equality can only be the fruit of education derived from the influence
of good laws." That is indeed the point. Education should have but one
object; to reduce the passions to equality, or rather to _equanimity_,
and to a certain equilibrium of mind. The education given to modern
democracy does not lead to this, but leads in the opposite direction.




CHAPTER XII.

THE DREAM.


What remedies can we apply to this modern disease, the worship of
intellectual and moral incompetence? What is, as M. Fouillée puts it,
the best way of avoiding the hidden rocks which threaten democracies? It
is hard to say, for we have to do with an evil which can only be cured
by itself, with an evil which is more than content with itself.

M. Fouillée (in the _Revue des Deux Mondes_ of November, 1909) proposes
an aristocratic Upper Chamber, that is to say, one that would represent
all the competence of the country, inasmuch as it would be appointed by
everything which is based on some particular form of excellence, the
magistracy, the army, the university, the chambers of commerce, and so
on.

Nothing could be better; but the consent of the democracy would be
necessary, and it is precisely these incorporations of efficiency that
the democracy cannot abide, looking on them, not without reason, as
being in a sense aristocracies.

He proposes also an energetic intervention on the part of the State to
restore public morality, action for the suppression of alcoholism,
gambling and pornography.

Beyond the fact that his argument savours of reaction, for it recalls to
us the programme of "moral order" of 1873, we must remark, as indeed M.
Fouillée himself acknowledges, that the democratic State can hardly
afford to kill the thing which enables it to live, to destroy its
principal source of revenue. Democracy, as its most authoritative
representatives have admitted, is not a cheap form of government. It has
always been instituted with the hope, and partly with the expressed
design, of being an economical government, and it has always been
ruinous, because it requires a much larger number of partisans than
other forms of government, and a smaller number of malcontents than
other forms of government, and these partisans have to be remunerated in
one fashion or another and the malcontents have to be silenced and
bought in one way or another.

Democracy, whether ancient or modern, lives always in terror of tyrants
who are always imminent or thought by it to be imminent. Against this
possible tyrant who would govern with an energetic minority, the
democracy requires an immense majority which it has to bind to it by the
grant of many favours; it has also to detach from this tyrant the
malcontents who would be his supporters if it did not disarm them by a
still more lavish distribution of favours.

Democracy requires therefore plenty of money. It will find this by
despoiling the wealthy as much as possible; but this is a very limited
source of revenue, for the wealthy are not a numerous class. It will
find it more easily, more abundantly also, by exploiting the vices of
all, for all is a very numerous group. Hence the complaisance shown to
drinking shops, which, as M. Fouillée remarks, it would be more
dangerous for the Government to close than to close the churches. As the
needs of the Government increase, as M. Fouillée predicts, without much
doubt it will claim a monopoly in houses of ill-fame and in the
publication of indecent literature; enterprises in which there would be
money. And after all, tolerating such things for the profit of certain
traders and annexing them to be worked for the profit of the State, is
surely much the same thing from a moral point of view. And the financial
operation would be much more beneficent in the second case than in the
first.

M. Fouillée also argues that reform must come "from above and not from
below," and that "the movement for regeneration can come from above and
not from below."

I ask nothing better, but I ask also how is it going to be done?
Inasmuch as everything depends upon the people, who, what, can influence
the people except the people itself? Everything depends on the people,
by what then can it be moved except by a force that is innate. We are
here confronted--we are talking to a philosopher and can make use of
scientific terms--with a {Kinêtês akinêtos} with a motive force
which causes but does not receive motives.

A principle has disappeared, a prejudice if you like to call it so, the
prejudice in favour of competence. We no longer think that the man who
understands how to do a thing ought to be doing that thing, or ought to
be chosen to do it. Hence, not only is everything mismanaged, but it
seems impossible by any device to handle the matter effectually. We see
no solution.

Nietzsche really has a horror of democracy; only like all energetic
pessimists, who are not mere triflers, he used to say from time to time:
"There are pessimists who are resigned and cowardly. We do not wish to
be like them." When he would not take this view he persuaded himself to
look at democracy through rose-coloured spectacles.

At times, looking at the matter from an æsthetic point of view, he used
to say: "Intercourse with the people is as indispensable and refreshing
as the contemplation of vigorous and healthy vegetation," and although
this is in flagrant contradiction to all he has elsewhere said of the
"bestial flock" and the "inhabitants of the swamp," the thought has a
certain amount of sense in it. It signifies that instinct is a force,
and that every force must be interesting to study; and further that, as
such, it contains an active virtue, a principle of life, a nucleus of
growth.

This, though vaguely expressed, is very possible. After all the crowd
is only powerful by reason of numbers, and because it has been decided
that numbers shall decide. It is an expedient; but an expedient cannot
impart force to a thing that had it not before. Motive power,
initiative, belongs to the man who has a plan, who makes his combination
to achieve it, who perseveres and is patient and does not relinquish
pursuit. If he is eliminated and reduced to impotence or to a minimum of
usefulness, one does not see how the crowd, without him, can obtain its
power of initiation. Further explanation is needed.

At another time, Nietzsche asks whether we ought not to respect the
right, which after all belongs to the multitude, to direct itself
according to an ideal--there are of course many ideals--and according to
the ideal which is its own. Ought we to refuse to the masses the right
to search out truth for themselves, the right to believe that they have
found it when they come upon a faith that seems to them vital, a faith
that is to them as their very life? The masses are the foundation on
which all humanity rests, the basis of all culture. Deprived of them,
what would become of the masters? It is to their interest that the
masses should be happy. Let us be patient; let us grant to our insurgent
slaves, our masters for the moment, the enjoyment of illusions which
seem favourable to them.

So Nietzsche argues, but more often, for he returns on various occasions
to this idea, led thereto by his customary aristocratic leanings, he
speaks of democracy as of a form of decadence, as a necessary prelude to
an aristocracy of the future. "A high civilisation can only be built
upon a wide expanse of territory, upon a healthy and firmly consolidated
mediocrity." [So he wrote in 1887. Ten years earlier he held that
slavery had been the necessary condition of the high civilisation of
Greece and Rome.] The only end, therefore, which at present,
provisionally of course but still for a long time to come, we have to
expect, must be the decadence of mankind--general decadence to a level
mediocrity, for it is necessary to have a wide foundation on which a
race of strong men can be reared. "The decadence of the European is the
great process which we cannot hinder, which we ought rather to
accelerate. It is the active cause at work which gives us hope of
seeing the rise of a stronger race, a race which will possess in
abundance those same qualities which are lacking to the degenerate
vanishing species, strength of will, responsibility, self-reliance, the
power of concentration...."

But how, out of this mediocrity of the crowd, a mediocrity which, as
Nietzsche says, is always increasing, by what process natural or
artificial can a new and superior race be created? Nietzsche seems to be
recalling the theory, very disrespectful and very devoid of filial
piety, by which Renan sought to explain his own genius. "A long line of
obscure ancestors," he says, "has economised for me a store of
intellectual energy," and he jots down in his note book certain
suggestions, a little immature but still emitting a ray of light. "It is
absurd," he says, "to imagine that this victory or survival of values
(that is low values, values, that is, that seem to be mediocrity) can be
antibiological: we must look for an explanation in the fact that they
are probably of some vital importance to the maintenance of the type
'man' in the event of its being threatened by a preponderance of the
feeble-minded and degenerate. Perhaps if things went otherwise, man
would now be an extinct animal. The elevation of type is dangerous for
the preservation of the species. Why? _Strong races are wasteful, we
find ourselves here confronted with a problem of economy._"

We perceive, in this train of reasoning, some inkling of what Nietzsche
is trying to formulate as his solution of the difficulty. What is needed
must be a natural process, a _vis medicatrix naturæ_. In the process of
declining and falling, races practise a sort of thrift; they save and
they economise. Then, if we may suppose that the quantity of energy of
intellectual and moral power, _i.e._, of "human values" at the disposal
of the race is constant, the races that so act are creating in
themselves a reserve which one day will irresistibly take shape in a
chosen class. They are creating in their own bosom an _élite_ which will
one day emerge, they have conceived all unconsciously an aristocracy
which will one day be born to be their ruler.

We always find in Nietzsche the theory of Schopenhauer, the theory of
the great deceiver who leads the human race by the nose and who makes
it do and, as if it liked it, that which it would never do if it knew
where it was being led. It is very possible; still it remains that
economy carried to an extreme, though it can lead to a reserve of force,
may also lead, and perhaps much more surely, to a condition of anæmia;
the annihilation of one set of competent people in order to prepare the
way for races of competent people in the future, I do not know if this
is a game inspired by the great deceiver, but it is a game which to me
appears dangerous. We ought to be sure (and who is sure?) that the great
deceiver does not abandon those who abandon themselves.

I have often said, without thinking of any metaphysical mythology,
thinking indeed of the ambitious people whom we meet everywhere, and
thinking only of giving them some good advice: "The best way to get
there is to come down." Nothing could be more philosophical, Nietzsche
would reply; it is even more true of peoples than of individuals: the
best way for peoples to become one day great is to begin by growing
smaller. I rather doubt it. There is no really solid reason to support
the theory that feebleness cultivated with perseverance results in
strength. Neither Greece nor Rome supply examples, nor did the
democratic republic of Athens nor the democratic Cæsarism of Rome ever
succeed in giving birth to an aristocracy of competence by a prolonged
economy of values.


--They did not have the time.--

Ah yes, there is always that to be said.


It would perhaps be better to try to put the brake on democracy than to
encourage this process of degeneration on the chance of a favourable
resurrection. At least this is the course which presents itself most
naturally to our mind, and which seems most consonant with duty.

When I say put the brake on democracy, it must be understood that I mean
that it should put the brake on itself, for nothing else can stop it,
when once it has made up its mind. It must be persuaded or left alone,
and even persuasion is a rash experiment, for it dislikes being
persuaded of anything but of its own omnipotence. It must be persuaded
or left alone, for every other method would be still more useless.

It must be reminded that forms of government perish from the abandonment
and also from the exaggeration of the principle from which their merit
is derived, though this is a very superannuated maxim; that they perish
by an abandonment of their principle because that principle is the
historical reason of their coming into existence, and they perish by
carrying their principle to excess, because there is no such thing as a
principle that is absolutely good and sufficient in itself for
regulating the complexity of the social machine.

What do we understand by the principle of a government? It is not that
which makes it be such and such a thing, but that "which makes it act"
in a particular way, as Montesquieu has remarked; that is, "the human
passions which supply the motive forces of life." It is clear then that
the passion for sovereignty, for equality, for incompetence, is not
sufficient to give to a government a life which is at once complete and
strong.

It is necessary to give to competence its part, or rather it is
necessary to give competence one part, for I do not wish to argue that
there is any question of right involved, I only affirm that it is a
social necessity. It is necessary that competence, technical,
intellectual, moral competence should be assigned its part to play, even
though the sovereignty of the people should be limited and the principle
of equality be somewhat abridged thereby.

A democratic element is essentially necessary to a people, an
aristocratic element also is essentially necessary to a people.

A democratic element is essentially necessary to a people in order that
the people should not feel itself to be a mere onlooker, but should
realise that it is a part and an important part of the body social, and
that the words "You are the nation, defend it," have a meaning.
Otherwise the argument of the anti-patriot demagogues would be just.
"What is the good of fighting for one set of masters against another
set, since it will make no difference, only a change of masters?"

A democratic element is required in the government of a people, because
it is very dangerous that the people should be an enigma. It is
necessary to know what it thinks, what it feels, what it suffers, what
it desires, what it fears, and what it hopes, and as this can only be
learnt from the people itself, it is necessary that it should have a
voice which can make itself heard.

This should be done in one way or another, either by a Chamber of its
own which should be endowed with great authority, or by the presence in
a single chamber of a considerable number of representatives of the
people, or by plebiscites constitutionally instituted as necessary for
the revision of the constitution and for laws of universal interest, or
by the liberty of the press and the liberty of association and public
meeting. This would not perhaps be enough, but it would be almost
enough. It is necessary that the people should be able to make known its
wants, and to influence the decisions of the Government, in a word its
voice should be heard and considered.

An aristocratic element is also necessary in a nation and in the
government of a nation so that all that admits of precision shall not be
smothered by that which is confused; so that what is exact shall not be
obscured by what is vague, and so that its firm resolves shall not be
shaken by vacillating and incoherent caprice.

Sometimes history itself makes an aristocracy--a fortunate circumstance
for a nation! This forms a caste more or less exclusive, it has
traditions, traditions more conservative of the laws than the laws
themselves, and it embodies in itself all that there is of life, and
energy and growth in the soul of a people. Sometimes history has failed
to give us an aristocracy or that which history has made has
disappeared. It is then that the people ought to draw one out of itself,
it is then its duty to appropriate and preserve the high qualities to be
found in men who have rendered service to the State or whose ancestors
have rendered service to the State, who have special qualifications for
each particular office and a moral efficiency for every form of public
service.

These qualities constitute the acquired aptitude of an aristocracy for
taking a part in the government; these qualities constitute its
adaptation to its social environment, and to its special function in our
social machinery and organisation. One might say that it is by these
qualities that _it enters into and becomes part of the organism of which
it is the material_. As John Stuart Mill has justly remarked, there
cannot be an expert, well-managed democracy if democracy will not allow
the expert to do the work which he alone can do.

What is wanted then and will always be wanted, even under socialism
where, as I pointed out, there will still be an aristocracy though a
more numerous one, is a blending of democracy and aristocracy; and here,
though he wrote a long time ago, we shall find Aristotle is always right
for he studied in a scientific spirit some hundred and fifty different
constitutions.

He is an aristocrat, without concealment, as we have seen, but his final
conclusions, whether he is speaking of Lacedæmon, which he did not like,
or of Carthage, or in general terms, have always been in favour of mixed
constitutions as ever the best. "There is," he says, "a manner of
combining democracy and aristocracy--which consists in so arranging
matters that both the distinguished citizens and the masses have what
they want. The right of every man to aspire to magisterial appointments
is a democratic principle, but the admission of distinguished citizens
only is an aristocratic principle."

This blending of democracy and aristocracy makes a good constitution,
but the union must not be one of mere juxtaposition which would serve
only to put hostile elements within striking distance. I said a
"blending" but the blending must be a real fusion. Our need is that in
the management of public business aristocracy and democracy should be
combined.

How? Well for many years I have been saying it and I hope I may live for
many years longer to say it again. A healthy nation is one in which the
aristocracy is "_demophil_," that is a lover of the people, and where
the people is aristocratic in its leanings. Every people where the
aristocracy is aristocratic and where the democracy is democratic is a
people destined to perish promptly, because it does not understand what
a people is, it has not got beyond the stage of knowing what is a class
and perhaps not even as far as that.

Montesquieu praises highly the Athenians and the Romans for the
following reason. "At Rome, although the people had the right of
elevating plebeians to office, it could never bring itself to elect
them; and although at Athens, it could by the law of Aristides, choose
magistrates from all classes, it never happened according to Xenophon,
that the lower people demanded the election of rulers who could injure
its safety and its glory. The two instances are identical; only, as far
as Athens is concerned, it signifies nothing, for at Athens everything
was decided by plebiscite and in consequence the real rulers of Athens
were the orators, in whom the people trusted, who enforced their
decisions and really governed the city. At Rome the fact is of great
importance for it was the elected magistrates who governed."

Republican Rome was indeed a country aristocratically governed which
had, however, a democratic element in its constitution, and this
democratic element, up to the time of the civil wars, was itself
profoundly aristocratic, just as the aristocracy which was always open
to an accession of members from the plebs was profoundly "demophil."

The institution of patron and client, even in the state of degeneracy
which overtook it, is a phenomenon which I believe is well-nigh unique.
It shows to what extent two classes felt the social necessity, the
patriotic necessity of mutual support and of a recognition of an
identity of interest.

A nation whose people is aristocratic and whose aristocracy is
"demophil" is a healthy nation. Rome succeeded in the world because for
five hundred years she enjoyed this social health.

An aristocratic people and a people-loving aristocracy. I had long
believed the formula was of my own invention. I have just discovered,
and I am in no way surprised, that Aristotle was before me. He quotes
the oath which oligarchs take in certain cities. "I swear to be always
the enemy of the people and never to counsel any thing that I do not
know to be injurious to them." "This," he continues, "is the very
opposite of what they ought to do or to pretend to do ... It is a
political fault which is often committed in oligarchies as well as in
democracies, and where the multitude has control of the laws, the
demagogues make this mistake. In their combat against the rich, they
always divide the State into two opposing parties. _In a democracy, on
the contrary, the Government should profess to speak for the rich, and
in oligarchies it should profess to speak in favour of the people._"

It is a Machiavelian counsel. Aristotle seems convinced that democrats
can only _profess_ to speak for the rich and that all we can expect from
oligarchs is an appearance of speaking in favour of the people.
Nevertheless he recognises clearly that for the peace and well-being of
the commonwealth such should be their attitude.

There is something more profound than this. Aristocrats ought not only
to appear but to be verily favourable to the demos, if they understand
the interests of aristocracy itself, for aristocracy requires a base.
Democrats also ought not only to appear but to be aristocratic if they
understand the interests of democracy which requires a guide.

This reciprocity of good offices, this reciprocity of devotion, and this
combination of effort are as necessary in modern as they were in ancient
republics. It is, and we must coin a word to express it, a social
"synergy" that is wanted. A union of all the vitalizing elements is as
necessary in society as in the family. Every family that is divided must
perish, every kingdom that is divided must perish.

I have said little of royalty which only indirectly concerns my subject.
If we have seen instances of the institution of royalty firmly
established, it is where the sentiment of royalty, appealing at once
both to the aristocracy and to the people, has realised that "synergy"
of the whole community of which we speak; it is where both, being united
in devotion to one object, are led to be devoted to each other by reason
of this convergence of their wills. _Eadem velle, eadem nolle amicitia
est._

There is no need of royalty for this. Royalty is our country itself
personified in one man. In the identification of country and kingdom, we
can and must arrive at this same union of the separate vitalities of the
nation, at this same community and convergence of will. The humble must
love their country in loving the great and the great must love their
country in loving the humble; and so all classes must be at one in their
hopes and in their fears. _Amicitia sit!_




INDEX.


 Abbeville, 115

 Abolition of Inheritance, 200, 203

 Academies, 27

 America, 170

 Amoeba, 15 _et seq._

 Antisthenes, 132

 Aristides, law of, 232

 Aristocracy, 36, 90, 137, 159
   ---- aptitude for government, 230
   ---- constitution which obeys laws, 88
   ---- demophil, 2, 232, 233
   ---- education under, 190
   ---- and examination system, 176
   ---- fusion with democracy, 231-233
   ---- impossible without merit, 31
   ---- old men under, 147
   ---- of Parliament, 172
   ---- permanent senators form, 58
   ---- and religion, 205, 207
   ---- result of indirect election, 19
   ----- and special jurisdictions, 98

 Aristophanes, 126, 150

 Aristotle, 14, 16, 31, 70, 78, 85, 88, 129, 131, 134-136, 153, 208, 215,
   231, 234, 235

 _Arrondissement_, 44
   ---- _scrutin d'_, 83

 Atheism, 207, 208

 Athens, 17, 31, 37, 83, 113, 149, 163, 226, 232

 Augier, 213

 Austria, 49


 Barthélemy, Abbé, 157

 Barthou, M., 9, 118

 Belgium, 170

 Bismarck, 209

 Bonald, 201

 Brunetière, 188

 Busiris, 115


 Calas, 115

 Carnot, 93

 Carthage, 16, 231

 Caucus, 6

 Chamber of Deputies, 44, 118, 172-174

 Charlemagne, 149

 Charondas, 87

 Church, the, 98, 206

 Cicero, 70
   ---- _de Senectute_, 149

 Civility, 156 _et seq._

 Civil Service, the, appointments to, 45-51
   ---- examinations for, 176 _et seq._

 Clergy, the, 169

 Code, the, 75, 76, 163

 Colbert, 43

 Collectivism, 200-203

 Communism, 200

 _Compétence par collation_, 19, 22

 Competitive examination, 175 _et seq._

 Conciliation Boards, 97

 _Concordat_, 206

 Constitution of 1873, 57
   ---- mixed, 231

 Co-optation, 176

 _Cour de Cassation_, 57, 101, 108, 176

 Court of Appeal, 57, 107

 Courts, ecclesiastical, 98
   ---- martial, 98

 Criminal procedure, 83
   ---- jurisdiction, 99


 Dandin, 184

 _Dandino-mania_, 184

 Decadence, 222, 223

 Decrees, 84, 91, 208, 209

 Demagogues, 85, 191, 213

 Democracy, Aristotle on, 129-136
   ---- Athenian, 17
   ---- children under, 142-146
   ---- and direct government, 37 _et seq._

 Democracy, encouragement of incompetence under, 30, 92, 126
   ---- English, 170
   ---- evolution of a modern, 18 _et seq._
   ---- fusion with aristocracy, 231-235
   ---- governed by decrees, 34, 91
   ---- and imperative mandate, 32
   ---- lack of respect under, 158-161
   ---- legislation under, 66, 79-81
   ---- magistrature, 110-122
   ---- Montesquieu on, 137-139
   ---- morality under, 139
   ---- Nietzsche on, 220-226
   ---- old men under, 146-153
   ---- Plato on, 127-129
   ---- and politicians, 34, 36
   ---- position of women under, 140-142
   ---- principle of, 14, 15
   ---- and private enterprise, 59-61
   ---- and reform, 216-219
   ---- Rousseau on, 136, 137
   ---- and schoolmasters, 191-215
   ---- and socialism, 61-65
   ---- and special jurisdictions, 98, 99

 Demos, 56, 86

 Demosthenes, 82

 Deputy, 55, 174
   ---- mayors, 119

 Despotism, principle of, 12
   ---- tendency of democracy towards, 62, 65, 96

 D'Etalonde, 115

 Diderot, 207

 Division of labour, 15
   ---- ---- in domestic life, 141

 Divorce, 125


 Germans, 67

 Gerontocracy, 146, 153

 Gladstone, Mr., 3, 5

 Goncourt, 201

 Greece, 222, 226

 Greeks, 71, 73, 87

 Greek philosophers, 126


 Holbach, 207

 Homer, 56

 Horace, 152

 Hugo, Victor, 209


 Ideal legislator, the, 79

 Imperative mandate, 32, 33

 Indirect election, 19

 Inequalities, artificial and natural, 30

 Interpellation, 83

 Italy, 170


 Japy, Mme., 114

 Jesuits in Paraguay, 62

 Joannès, Baron, 68

 Judges, appointment of, 46, 100-112
   ---- interpretation of the law, 163

 _Juge de paix_, 111, 116 _et seq._

 Jurisdiction, criminal, 99
   ---- ecclesiastical, 97
   ---- military, 97
   ---- seignorial, 97

 Jury, the 112-115, 168, 169


 Kant, 153


 La Barre, 115

 Lacedæmon, 150, 231

 Law, abolishing primogeniture, 124
   ---- abuse of, 110
   ---- of competence, 97
   ---- and decrees, 84-91
   ---- degree in, 116
   ---- doctors of, 108
   ---- ecclesiastical, 125
   ---- emergency, 82-89
   ---- fundamental, 89
   ---- governs men, 121
   ---- of July 12th, 1905, 117
   ---- made by the ideal legislator, 67-74
   ---- made by the people, 17, 18, 25, 40, 194-196, 234
   ---- must be ancient and unchanged, 77-79
   ---- not the same for rich and poor, 199
   ---- permitting divorce, 125
   ---- of proportion, 14
   ---- questions of, 113, 162
   ---- of Sunday observance, 75, 76
   ---- and tradition, 230

 Legal profession, the, 60

 Legislation, ancient, 87
   ---- English, 2, 4
   ---- party, 81-82
   ---- philanthropic, 6
   ---- predatory, 8
   ---- requires special knowledge, 18, 67

 Legislator, essential qualifications for, 66-81
   ---- Greek, 73

 Lestranger, M. Marcel, 164, 166

 Liberty of association, 194, 196

 Louis XIV, 21, 22, 150, 195, 204, 209
   ----    XV, 150
   ----    XVI, 133

 Louvois, 43

 Lycurgus, 2, 73, 87

 Lynching, 168, 169


 Machiavel, 71

 Machiavelian counsel, 234

 Magistracy, hereditary system advocated for, 105, 163, 168

 Magistrates, election of, 48, 100-112
   ---- incompetence of, 164-168
   ---- subservience to government, 163

 Mayors, 117

 Medical examinations, 180
   ---- profession, 60

 Mill, John Stuart, 230

 Minister of Agriculture, 94
   ---- of Commerce, 92
   ---- of Education, 92, 211

 Minister of Interior, 211
   ---- of Justice, 76, 118-120
   ---- of Navy, 92
   ---- of War, 92, 93

 Miscarriage of justice, 165

 Monarchy, old men under, 147
   ---- principle of, 12, 62, 96

 Montesquieu, 4, 12, 14, 28, 37, 44-47, 66-70, 74-89, 103-105, 123, 137,
 138, 150-156, 193, 227, 232

 Moral effect of law, 125
   ---- order, 217

 Morals, high standard of, 150
   ---- laxity of, 155
   ---- private, 139, 140
   ---- public, 123, 124, 139

 More, Thomas, 71

 Moses, 67, 209


 Napoleon, 13, 105, 209

 Nationalisation, 60

 Nietzsche, 147, 187, 220-225


 Oligarchy, 63, 128, 147

 Olympiad, 87

 Ostracism, 133


 Paraguay, 62

 Parental authority, 143

 Parliament, 172

 Party system under Socialism, 61

 Pascal, 107, 157

 Patriotism, 12, 214

 Peel, Sir R., 3

 Periander, 132

 Peter the Great, 124

 Phædo, personification of laws in, 87

 Plato, 71, 105, 126-129, 138, 140, 153

 Plautus, 150

 Plebiscite, 24, 173

 _Pli Professionnel, Le_, 164

 Pnyx, the, 37

 Politeness, 156 _et seq._

 Politician, definition of, 34
   ---- democracy's need of, 35, 51
   ---- as schoolmaster, 212

 _Préfet_, 62-55, 173, 212

 President of the Chamber, 110
   ---- of the Council, 93
   ---- of the Republic, 42, 57

 Primogeniture, 124

 _Procureur-général_, 52, 53, 119-122

 Proudhon, 73

 Public officials, appointment of, 43-51
   ---- under Socialism, 201

 Purchase system, the, 103


 Quinet, Edgar, 13


 Rabelais, 115

 Republic, 12, 13, 73

 Renaissance, 149

 Renan, 223

 Renouvier, 50

 Revocation of the Edict of Nantes, 22

 Rhédi, 78

 Robespierre, 207, 208

 Roman Republic, 13, 113, 204, 233

 Romans, the, 68, 87, 232

 Rome, 124, 149, 222, 226, 232-234

 Rousseau, 32, 72, 136-138, 149, 150, 207, 208

 Royalty, 137, 286
   ---- democratic, 18


 St. Cyr, 177

 St. Louis, 74

 Schopenhauer, 224

 _Scrutin de liste_, 83, 112, and note
   ---- _d'arrondissement_, 83

 Senate, 38, 58, 117, 118, 172-175

 Senators, 55, 172-175
   ---- irremovable, 57

 Seneca, 129

 Socialism, 61-65, 231

 Socrates, 18, 88
   ---- on democracy, 144

 Solon, 67, 73

 Sovereignty of the People, 40, 45, 56, 177, 194-198, 204-209, 214

 Specialisation of functions, 15-18, 31

 Special jurisdictions, 97, 98

 State control of children, 144
   ---- ---- of magistrates, 100,101
   ---- ---- of private enterprise, 60-61
   ---- danger of eminent man to, 131-137
   ---- intervention to restore public morality, 217-219

 State policy, 110
   ---- services rendered to, 230
   ---- within a State, 196

 Steinheil case, 114

 Suidas, 103, 105

 Sunday observance, 75

 Sylla, 73

 Synergy, 235, 236


 Terence, 150

 Thrasybulus, 132

 Toulouse, 115

 Tyranny, 135


 Universal suffrage, 21, 173-175

 University degrees, 20, 181-183

 Upper chamber, suggestions for strong, 174

 Usbek, 78


 Virtue, civic, 14, 103, 104
   ---- the principle of a republic, 12, 13

 Voltaire, 103-105, 116


 Women, their position in a democracy, 140-143


 Xenophon, 126, 233


 Young man, Horace's description of, 152




_Printed by Sherratt & Hughes, London and Manchester._





End of Project Gutenberg's The Cult of Incompetence, by Emile Faguet