Chapter III. Rationalization of Non-logical Conduct

249. The research just completed has called our attention—along with a number of incidental inductions—to the following facts:

1. The existence and importance of non-logical conduct. That runs counter to many sociological theories that either scorn or ignore non-logical actions, or else, in an effort to reduce all conduct to logic, attach little importance to them. The course we follow in studying the behaviour of human beings as bearing on the social equilibrium differs according as we lay the greater stress on logical or non-logical conduct. We had better look into that matter more deeply, therefore.

2. Non-logical actions are generally considered from the logical standpoint both by those who perform them and by those who discuss them and generalize about them. Hence our need to do a thing of supreme importance for our purposes here—to tear off the masks non-logical conduct is made to wear and lay bare the things they hide from view. That too runs counter to many theories which halt at logical exteriors, representing them not as masks but as the substantial element in conduct itself. We have to scrutinize those theories closely: for if we were to find them true—in accord with experience, that is—we would have to follow an altogether different course from the one we would follow were we to discover that the substantial element in the conduct lies in the things that underlie the logical exteriors (§ 146).

3. The experimental truth of a theory and its social utility are different things. A theory that is experimentally true may be now advantageous, now detrimental, to society; and the same applies to a theory that is experimentally false. Many many people deny that. We must therefore not rest satisfied with the rapid survey we have made so far, much less with the bald declaration in § 72. We must see whether observation of the facts confirms or belies our induction (§§ 72-73).

4. As regards logical and non-logical conduct there are differences between individual human beings, or, taking things in the mass, between social classes, and differences also in the degrees of utility that theories experimentally true or experimentally false have for individuals or classes. And the same applies to the sentiments that are expressed through non-logical conduct. Many people deny such differences. To not a few the mere suggestion that they exist seems scandalous. It will therefore be necessary to continue our examination of that subject, on which we have barely touched, and clearly establish just what the facts have to say.

250. Meantime, our first survey has already given us an idea, however superficial, of the answers that have to be given to the inquiries suggested in §§ 13-14 as to the motives underlying theories, as to their bearing on experimental realities, and as to individual and group utilities—and we see that some at least of the distinctions that are drawn in those paragraphs are not merely hypothetical, but have points of correspondence with reality.

251. In the following pages we shall devote ourselves chiefly to running down non-logical actions in the theories or descriptions of social facts that have been put forward by this or that writer; and that will give us an approximate notion of the way non-logical conduct is masked by logic.

252. If non-logical actions are really as important as our induction so far would lead us to suppose, it would be strange indeed that the many men of talent who have applied themselves to the study of human societies should not have noticed them in any way. Distracted by preconceptions or led astray by erroneous theories, they may, "as they that have spent eyes," have caught imperfect glimpses of them; but it is hard to believe that they can have seen nothing where we find so much that is of such great significance. Let us therefore see just how the matter stands.

253. But for that purpose we have to take an even more general view of things: we have to see to what extent reality is disfigured in the theories and descriptions of it that one finds in the literature of thought. We have an image in a curved mirror; our problem is to discover the form of the object so altered by refraction.

Suppose we ignore, for the moment, the simplest case of writers who understand that the conduct of human beings depends, to some extent at least, on the environment in which they live, on climate, race, occupation, "temperament." It is obvious that the behaviour resulting from such causes is not the product of pure ratiocination, that it is non-logical behaviour. To be sure, that fact is often overlooked by the very writers who have stressed it, and they therefore seem to be contradicting themselves. But the inconsistency is now and again more apparent than real; for when a writer admits such causes he is usually dealing with what is—and that is one thing. When he insists on having all conduct logical, he is usually describing what, in his opinion, ought to be—and that is quite another thing. From the scientific laboratory he steps over into the pulpit.

254. Let us begin with cases not quite so simple but where it is still easy to perceive the experimental truth underneath imperfect and partly erroneous descriptions of it.

Here, for instance, is The Ancient City of Fustel de Coulanges. In it we read, p. 73 (Small, p. 89): "From all these beliefs, all these customs, all these laws, it clearly results that from the religion of the hearth human beings learned to appropriate the soil and on it based their title to it." But, really, is it not surprising that domestic religion should have preceded ownership of land? And Fustel gives ho proof whatever of such a thing! The opposite may very well have been the case—or religion and ownership of land may have developed side by side. It is evident that Fustel has the preconceived notion that possession has to have a "cause." On that assumption, he seeks the cause and finds it in religion; and so the act of possession becomes a logical action derived from religion, which in its turn can now be logically derived from some other cause. By a singular coincidence it happens that in this instance Fustel himself supplies the necessary rectification. A little earlier, p. 63 (Small, p. 78), he writes: "There are three things which, from the most ancient times, one finds founded and solidly established in these Greek and Italian communities: domestic religion, the family, the right of property— three things which were obviously related in the beginning and which seem to have been inseparable."

How did Fustel fail to see that his two passages were contradictory? If three things A, B, C are "inseparable," one of them, for instance A, cannot have produced another, for instance B: for if A produced B, that would mean that, at the time, A was separate from B. We are therefore compelled to make a choice between the two propositions. If we keep the first, we have to discard the second, and vice versa. As a matter of fact, we have to adopt the second, discarding the proposition that places religion and property in a relationship of cause and effect, and keeping the one that puts them in a relationship of interdependence (§§ 138, 267). The very facts noted by Fustel himself force that choice upon us. He writes, p. 64 (Small, p. 79): "And the family, which by duty and religion remains grouped around its altar, becomes fixed to the soil like the altar itself." But the criticism occurs to one of its own accord: "Yes, provided that be possible!" For if we assume a social state in which the family cannot settle on the soil, it is the religion that has to be modified. What obviously has happened is a series of actions and reactions, and we are in no position to say just how things stood in the beginning. The fact that certain people came to live in separate families fixed to the soil had as one of its manifestations a certain kind of religion; and that religion, in its turn, contributed to keeping the families separate and fixed to the soil (§ 1021).

255. In this we have an example of a very common error, which lies in substituting relationships of cause and effect for relationships of interdependence (§ 138); and that error gives rise to still another: the error of placing the alleged effect, erroneously regarded as the logical product of the alleged cause, in the class of logical actions.

256. When Polybius stresses religion as one of the causes of the power of Rome (§ 313), we will accept the remark as very suggestive; but we will reject the logical explanation that he gives of the fact (§ 313¹)

In Sumner Maine's Ancient haw, p. 122, we find another example like Fustel's. Maine observes that ancient societies were made up of families. That is a question of fact which we choose not to go into— researches into origins are largely hypothetical anyway. Let us accept Maine's data for what they are worth—just as hypotheses. From the fact he draws the conclusion that ancient law was "adjusted to a system of small independent corporations." That too is good: institutions adjust themselves to states of fact! But then suddenly we find the notion of logical conduct creeping stealthily in, p. 177: "Men are regarded and treated, not as individuals, but always as members of a particular group." It would be more exact to say that men are that in reality, and law, accordingly, develops as if men were regarded and treated as members of a particular group.

A little earlier, Maine's intromission of logical conduct is more obtrusive. Following his remark that ancient societies were made up of small independent corporations, he adds, p. 122: "Corporations never die, and accordingly primitive law considers the entities with which it deals, i.e., patriarchal or family groups, as perpetual and inextinguishable." From that Maine derives as a consequence the institution of transmission, upon decease, of the universitas iuris, which we find in Roman law. Such a logical sequence may easily be compatible with a posterior logical analysis of antecedent non-logical actions, but it does not picture the facts accurately. To come nearer to them we have to invert some of the terms in Maine's previous remarks. The succession of the universitas iuris does not derive from the concept of a continuous corporation: the latter concept derives from the fact of succession. A family, or some other ethnic group, occupies a piece of land, comes to own flocks, and so on. The fact of perpetuity of occupation, of possession, is in all probability antecedent to any abstract concept, to any concept of a law of inheritance. That is observable even in animals. The great felines occupy certain hunting-grounds and these remain properties of the various families, unless human beings chance to interfere.¹ The ant-hill is perpetual, yet one may doubt whether ants have any concept of the corporation or of inheritance. In human beings, the fact gave rise to the concept. Then man, being a logical animal, had to discover the "why" of the fact; and among the many explanations he imagined, he may well have hit upon the one suggested by Sumner Maine.

Maine is one of the writers who have best shown the difference between customary law (law as fact) and positive law (law as theory); yet he forgets that distinction time and again, so persuasive is the concept that posits logical conduct everywhere. Customary law is made up of a complex of non-logical actions that regularly recur. Positive law comprises two elements: first, a logical—or pseudo-logical or even imaginary—analysis of the non-logical actions in question; second, implications of the principles resulting from that analysis. Customary law is not merely primitive: it goes hand in hand with positive law, creeps unobtrusively into jurisprudence, and modifies it. Then the day comes when the theory of such modifications is formulated—the caterpillar becomes a butterfly—and positive law opens a new chapter.

257. Of the assassination of Caesar, Duruy writes:¹ "Ever since the foundation of the Republic the Roman aristocracy had adroitly fostered in the people a horror for the name of king." In that the logical varnish for conduct that is non-logical is easily recognizable. Then he goes on: "If the monarchical solution answered the needs of the times, it was almost inevitable that the first monarch should pay for his throne with his life, as our Henry IV paid for his." In such "needs of the times" we recognize at once one of those amiable fictions which historians try to palm off as something concrete. As for the law that first monarchs in dynasties have to die by assassination, history gives no experimental proof of any such fact. We have to see in it a mere reminiscence of the classical fatum, and pack it off to keep company with many similar products of the scholarly fancy.

258. Shall we banish from history the prodigies that Suetonius never forgets to enumerate in connexion with the births or deaths of the Roman Emperors, without trying to interpret them—for we shall see how mistaken such an effort on our part would be (§ 672)—and shall we keep only such of his facts as are, or at least seem to be, historical? Shall we do the same with all similar historical sources—for instance, with histories of the Crusades?¹

In doing that we should be on dangerous ground, for if we made it an absolute rule to divide all our narrative sources into two elements, one miraculous, incredible, which we reject, and another natural, plausible, which we retain, we should certainly fall into very serious errors (§ 674). The part that is accepted has to have extrinsic probabilities of truth, whether through the demonstrable credibility of the author or through accord with other evidence.

259. From a legend we can learn nothing that is strictly historical; but we can learn something, and often a great deal, about the psychic state of the people who invented or believed it; and on knowledge of such psychci states our research is based. We shall therefore often cite facts without trying to ascertain whether they are historical or legendary; because for the use we are going to make of them they are just as serviceable in the one case as in the other—sometimes, indeed, they are better legendary than historical.

260. Logical interpretations of non-logical conduct become in their turn causes of logical conduct and sometimes even of non-logical conduct; and they have to be reckoned with in determining the social equilibrium. From that standpoint, the interpretations of plain people are generally of greater importance than the interpretations of scholars. As regards the social equilibrium, it is of far greater moment to know what the plain man understands by "virtue" than to know what philosophers think about it.

261. Rare the writer who fails to take any account of non-logical conduct whatever; but generally the interest is in certain natural inclinations of temperament, which, willynilly, the writer has to credit to human beings. But the eclipse of logic is of short duration— driven off at one point, it reappears at some other. The rôle of temperament is reduced to lowest terms, and it is assumed that people draw logical inferences from it and act in accordance with them.

262. So much for the general situation. But in the particular, theorists have another very powerful motive for preferring to think of non-logical conduct as logical. If we assume that certain conduct is logical, it is much easier to formulate a theory about it than it is when we take it as non-logical. We all have handy in our minds the tool for producing logical inferences, and nothing else is needed. Whereas in order to organize a theory of non-logical conduct we have to consider hosts and hosts of facts, ever extending the scope of our researches in space and in time, and ever standing on our guard lest we be led into error by imperfect documents. In short, for the person who would frame such a theory, it is a long and difficult task to find outside himself materials that his mind supplied directly with the aid of mere logic when he was dealing with logical conduct.

263. If the science of political economy has advanced much farther than sociology, that is chiefly because it deals with logical conduct.¹ It would have been a soundly constituted science from the start had it not encountered a grave obstacle in the interdependence of the phenomena it examines, and at a time when the scholars who were devoting themselves to it were unable to utilize the one method so far discovered for dealing with interdependencies. The obstacle was surmounted, in part at least, when mathematics came to be applied to economic phenomena, whereby the new science of mathematical economics was built up, a science well able to hold its own with the other natural sciences.²

264. Other considerations tend to keep thinkers from the field of non-logical conduct and carry them over into the field of the logical. Most scholars are not satisfied with discovering what is. They are anxious to know, and even more anxious to explain to others, what ought to be. In that sort of research, logic reigns supreme; and so the moment they catch sight of conduct that is non-logical, instead of going ahead along that road they turn aside, often seem to forget its existence, at any rate generally ignore it, and beat the well-worn path that leads to logical conduct.

265. Some writers likewise rid themselves of non-logical actions by regarding them—often without saying as much explicitly—as scandalous things, or at least as irrelevant things, which should have no place in a well-ordered society. They think of them as "superstitions" that ought to be extirpated by the exercise of intelligence. Nobody, in practice, acts on the assumption that the physical and the moral constitution of an individual do not have at least some small share in determining his behaviour. But when it comes to framing a theory, it is held that the human being ought to act rationally, and writers deliberately close their minds to things that the experience of every day holds up before their eyes.

266. The imperfection of ordinary language from the scientific standpoint also contributes to the wide-spread resort to logical interpretations of non-logical conduct.

267. It plays no small part in the common misapprehension whereby two phenomena are placed in a relationship of cause and effect for the simple reason that they are found in company. We have already alluded to that error (§ 255); but we must now advance a little farther in our study of it, for it is of no mean importance to sociology.

Let C, as in Figure 3, § 166, stand for a belief; D, for certain behaviour. Instead of saying simply, "Some people do D and believe C," ordinary speech goes farther and says, "Some people do D because they believe C." Taken strictly, that proposition is often false. Less often false is the proposition, "Some people believe C because they do D." But there are still many occasions when all that we can say is, "Some men do D and believe C."

In the proposition, "Some people do D because they believe C," the logical strictness of the term "because" can be so attenuated that no relationship of cause and effect is set up between C and D. We can then say, "We may assume that certain people do D because they have a belief C which expresses sentiments that impel them to do D"; that is because (going back to Figure 3), they have a psychic state A that is expressed by C. In such a form the proposition closely approximates the truth, as we saw in § 166.

268. Figure 3 can be broken up into three others (Figure 7).

I. The psychic state A produces the belief C and the conduct D, there being no direct relation between C and D. That is the situation in the proposition, "People do D and believe C."

II. The psychic state A gives rise to the conduct D, and they both produce the belief C. That is the situation in the second proposition, "People believe C because they do D."

III. The psychic state A gives rise to the belief C, which produces the behaviour D. That is the situation in the proposition, "People do D because they believe C."

269. Although case III is not the only case, nor even the most frequent case, people are inclined to regard it as general and to merge with it cases I and II to which they preferably attribute little or no importance. Ordinary language, with its lack of exactness, encourages the error, because a person may state case III explicitly and be unconsciously thinking meantime of cases I and II. It often happens, besides, that we get mixtures of the three cases in varying proportions.

270. Aristotle opens his Politics, I, 1, 1 (Rackham, p. 3), with the statement: "Seeing that every city is a society (Rackham, "partnership") and that every society (partnership) is constituted to the end of some good (for all men work to achieve what to them seems good) it is manifest that all societies (partnerships) seek some good." Here we stand altogether in the domain of logic: with a deliberate purpose—the purpose of achieving a certain good—human beings have constituted a society that is called a city. It would seem as though Aristotle were on the point of going off into the absurdities of the "social contract"! But not so. He at once changes tack, and the principle he has stated he will use to determine what a city ought to be rather than what it actually is.

271. The moment Aristotle has announced his principle—an association for purposes of mutual advantage—he tosses it aside and gives an altogether different account of the origin of society. First he notes the necessity of a union between the sexes, and soundly remarks that "that does not take place of deliberate choice"¹; wherewith, evidently, we enter the domain of non-logical conduct. He continues: "Nature has created certain individuals to command and others to obey." Among the Greeks Nature has so distinguished women and slaves. Not so among the Barbarians, for among the Barbarians, Nature has not appointed any individuals to command. We are still, therefore, in the domain of non-logical conduct; nor do we leave it when Aristotle explains that the two associations of master and slave, husband and wife, are the foundations of the family, that the village is constituted by several families, and that several villages form a state; nor when, finally, he concludes with the explicit declaration that "Every city, therefore, like the original associations, comes of Nature."² One could not allude to non-logical actions in clearer terms.

272. But, alas, if the city comes of Nature, it does not come of the deliberate will of citizens who get together for the purpose of achieving a certain advantage! There is an inconsistency between the principle first posited and the conclusion reached.¹ Just how Aristotle fell into it we cannot know, but to accomplish that feat for oneself, one may proceed in the following fashion: First centre exclusively on the idea of "city," or "state." It will then be easy to connect city, or state, with the idea of "association," and then to connect association with the idea of deliberate association. So we get the first principle. But now think, in the second place, of the many many facts observable in a city or a state—the family, masters and slaves, and so on. Deliberate purpose will not fit in with those things very well. They suggest rather the notion of something that develops naturally. And so we get Aristotle's second description.

273. He gets rid of the contradiction by metaphysics, which never withholds its aid in these desperate cases. Recognizing non-logical conduct, he says, I, 1, 12 (Rackham, pp. 11-13): "It is therefore manifest that the city is a product of Nature and is superior (prior) to man (to the individual). From Nature accordingly comes the tendency (an impulse) in all men toward such association. Therefore the man who first founded one was the cause of a very great good." So then, there is the inclination imparted by Nature; but it is further necessary that a man found the city. So a logical action is grafted upon the non-logical action (§ 306, I-β); and there is no help for that, for, says Aristotle, Nature does nothing in vain.¹ Our best thanks, therefore, to that estimable demoiselle for so neatly rescuing a philosopher from a predicament!

274. In distinguishing the Greeks from the Barbarians in his celebrated theory of natural slavery, Aristotle avails himself of the concept of non-logical conduct. It is obvious, among other things, that logic being the same for Greeks and Barbarians, if all actions were logical there could not be any difference between Greeks and Barbarians. But that is not all. Good observer that he is, Aristotle notices differences among Greek citizens. Speaking of the forms of democracy he says, VI, 2, 1 (Rackham, pp. 497-98): "Excellent is an agricultural people; consequently one can institute a democracy where a people lives by farming and sheep-raising." And he repeats, VI, 2, 7 (Rackham, p. 503): "Next after farmers, the best people are shepherds, or people who live by owning cattle. . . . The other rabbles on which other sorts of democracy are based are greatly inferior." Here then we get clearly distinguished classes of citizens and almost a rudimentary economic determinism. But there is no reason for our stopping where Aristotle stops; and if we do go on we see that in general the conduct of human beings depends on their temperaments and occupations.

Cicero credits the ancestors of the Romans of his time with knowing that "the characters of human beings result not so much from race and family as from those things which are contributed by the nature of their localities for the ordinary conduct of life, and from which we draw our livelihood and subsistence. The Carthaginians were liars and cheats not by race but from the nature of their country, which with its port and its contacts with all sorts of merchants and foreigners speaking different languages inclined them through love of profits to love of trickery. The mountaineers of Liguria are harsh and uncouth. . . . The Capuans have ever been a supercilious people, because of the fertility of their soil, the wealth of their harvests, the salubriousness, the disposition, and the beauty of their city."¹

275. In his Rhetoric, II, 12-14 (Freese, pp. 247-57), Aristotle makes an analysis, which came to be celebrated, of the traits of man according to age—in adolescence, in maturity, and in senility. He pushes his analysis further still, II, 12, 17 (Freese, pp. 257-63), and examines the effects on character of noble birth, wealth, and power—a splendidly conducted study. But all that evidently carries him into the domain of non-logical conduct.¹

276. Aristotle even has the concept of evolution. In the Politics, II, 5, 12 (Rackham, pp. 129-31), he remarks that the ancestors of the Greeks probably resembled the vulgar and ignorant among his contemporaries.

277. Had Aristotle held to the course he in part so admirably followed, we would have had a scientific sociology in his early day. Why did he not do so? There may have been many reasons; but chief among them, probably, was that eagerness for premature practical applications which is ever obstructing the progress of science, along with a mania for preaching to people as to what they ought to do—an exceedingly bootless occupation—instead of finding out what they actually do. His History of Animals avoids those causes of error, and that perhaps is why it is far superior to the Politics from the scientific point of view.

278. It might seem strange to find traces of the concept of non-logical conduct in a dreamer like Plato; yet there they are! The notion transpires in the reasons Plato gives for establishing his colony far from the sea. To be near the sea begins by "being sweet" but ends by "being bitter" for a city: "for filling with commerce and traffic it develops capricious, untrustworthy instincts, and a breed of tricksters."¹ Non-logical conduct has its place also in the well-known apologue of Plato on the races of mankind. The god who fashioned men mixed gold into the composition of those fit to govern, silver in guardians of the state (the warriors), iron in tillers of the soil and labourers. Plato also has a vague notion of what we are to call class-circulation, or circulation of élites (§§ 2026 f.). He knows that individuals of the silver race may chance to be born in the race of gold, or vice versa, and so for the other races.²

279. That being the case, if one would remain within the domain of science, one must go on and investigate the probable characteristics and the probable evolution of a society made up of different races of human beings, which are not reproduced from generation to generation with exactly the same characteristics and which are able to mix. That would be working towards a science of societies. But Plato has a very different purpose. He is little concerned with what is. He strains all his intellectual capacities to discover what ought to be. And thereupon non-logical conduct vanishes, and Plato's fancy goes sporting about among logical actions, which he invents in great numbers; and we find him at no great cost to himself appointing magistrates to put individuals who are born in a class but differ in traits from their parents in their proper places, and proclaiming laws to preserve or alter morals—in short, deserting the modest province of science to rise to the sublime heights of creation.

280. The controversies on the question "Can virtue be taught?" also betray some distant conception of non-logical conduct. According to the documents in our possession, it would seem that Socrates regarded virtue as a science and left little room for non-logical actions.¹ Plato and Aristotle abandon that extreme position. They hold that a certain natural inclination is necessary to "virtue." But that inclination once premised, back they go to the domain of logic, which is now called in to state the logical implications of temperament, and these in their turn determine human conduct. Those old controversies have points of resemblance with the disputes which took place long afterwards on "efficacious," and "non-efficacious," grace.

281. The procedure of Plato and Aristotle in the controversies on the teaching of "virtue" is a general one. Non-logical actions are credited with a rôle that it would be absurd not to give them, but then that rôle is at once withdrawn, and people go back to the logical implications of inclinations; and by dividing those inclinations, which in fact cannot be ignored, into "good" ones and "bad" ones, a way is found to keep inclinations that are in accord with the logical system one prefers and to eliminate all others.

282. St. Thomas tries to steer a deft course between the necessity of recognizing certain non-logical inclinations and a great desire to give full sway to reason, between the determinism of non-logical conduct and the doctrine of free will that is implicit in logical conduct. He says that "virtue is a good quality or disposition (habitus) of the soul, whereby one lives uprightly, which no one uses wrongly, and which God produces within us apart from any action by ourselves."¹ Taken as a "disposition of the soul" virtue is classed with non-logical actions; and so it is when we say that God produces it in us apart from anything we do of ourselves. But by that divine interposition any uncertainty as to the character of non-logical conduct is removed, for it becomes logical according to the mind of God and therefore logical for the theologians who are so fortunate as to know the divine mind. Others use Nature for the same purpose and with the same results. People act according to certain inclinations. That reduces the rôle of the non-logical to a minimum, actions being regarded as logical consequences of the inclinations. Then even that very modest remnant is made to vanish as by sleight-of-hand; for inclinations are conceived as imparted by some entity (God, Nature, or something else) that acts logically (§ 306, I-β); so that even though the acting subject may on occasion believe that his actions are non-logical, those who know the mind, or the logical procedure, of the entity in question—and all philosophers, sociologists, and the like, have that privilege—know that all conduct is logical.

283. The controversy between Herbert Spencer and Auguste Comte brings out a number of interesting aspects of non-logical conduct.

284. In his Lectures on Positive Philosophy (Cours de philosophie positive) Comte seems to be decidedly inclined to ascribe the predominance to logical conduct. He sees in positive philosophy, Vol. I, pp. 48-49, "the one solid basis for that social reorganization which is to terminate the critical state in which civilized nations have been living for so long a time." So then it is the business of theory to reorganize the world! How is that to come about? "Not to readers of these lectures should I ever think it necessary to prove that ideas govern and upset the world, or, in other terms, that the whole social mechanism rests, at bottom, on opinions. They are acutely aware that the great political and moral crisis in present-day society is due, in the last analysis, to our intellectual anarchy. Our most serious distress is caused by the profound differences of opinion that at present exist among all minds as to all those fundamental maxims the stability of which is the prime requisite for a real social order. So long as individual minds fail to give unanimous assent to a certain number of general ideas capable of constituting a common social doctrine, we cannot blind ourselves to the fact that the nations will necessarily remain in an essentially revolutionary atmosphere. . . . It is just as certain that if this gathering of minds to one communion of principles can once be attained, the appropriate institutions will necessarily take shape from it."

285. After quoting Comte's dictum that ideas govern and upset the world, Herbert Spencer advances a theory that non-logical actions alone influence society. "Ideas do not govern and overthrow the world: the world is governed or overthrown by feelings, to which ideas serve only as guides. The social mechanism does not rest finally upon opinions; but almost wholly upon character. Not intellectual anarchy, but moral antagonism, is the cause of political crises. All social phenomena are produced by the totality of human emotions and beliefs. . . . Practically, the popular character, and the social state, determine what ideas shall be current; instead of the current ideas determining the social state and the character. The modification of men's moral natures caused by the continuous discipline of social life, which adapts them more and more to social relations, is therefore the chief proximate cause of social progress."¹

286. Then a curious thing happens: Comte and Spencer reverse positions reciprocally! In his System of Positive Polity, Vol. IV, p. 5, Comte decides to allow sentiment to prevail, and expresses himself very clearly on the point: "Though I have always proclaimed the universal preponderance of sentiment, I have had, so far, to devote my attention primarily to intelligence and activity, which prevail in sociology. But the very real ascendancy they have acquired having now brought on the period of their real systematization, the final purpose of this volume must now be to bring about a definite predominance of sentiment, which is the essential domain of morality."

Comte is straining the truth a little when he says that he has "always proclaimed the universal preponderance of sentiment." No trace of any such preponderance is to be detected in his Cours. Ideas stand in the forefront there. But Comte has changed. He began by considering existing theories, which he wished to replace with others of his own make; and in that battle of ideas, his own naturally won the palm, and from them new life was to come to the world. But time rolls on. Comte becomes a prophet. The battle of ideas is over. He imagines he has won a complete victory. So now he begins proclaiming dogma, pronouncing ex cathedra, and it is only natural that nothing but sentiments should now be left on the field—his own sentiments, of course.¹

287. Comte, moreover, began by hoping to make converts of people; and naturally the instrument for doing that was, at the time, ideas. But he ended by having no hope save in a religion imposed by force, imposed if need be by Czar Nicholas, by the Sultan, or at the very least by a Louis Napoleon (who would in fact have done better to rest content with being just a dictator in the service of Positivism).¹ In this scheme sentiment is the big thing beyond shadow of doubt, and one can no longer say that "ideas govern or upset the world." It would be absurd to suppose that Comte turned to the Czar, to Reshid Pasha, or to Louis Napoleon, to induce them merely to preach ideas to their peoples. One might only object that the ideas of Comte would be determining the religion which would later be imposed upon mankind; and in that case ideas would be "upsetting the world," if the Czar, the Sultan, Louis Napoleon, or some other well-intentioned despot saw fit to take charge of enforcing Comte's positivism upon mankind. But that is far from being the meaning one gathers from the statements in the Cours.

288. Comte recognizes, in fact he greatly exaggerates, the social influence of public worship and its efficacy in education—all of which is just a particular case of the efficacy of non-logical impulses. If Comte could have rested satisfied with being just a scientist, he might have written an excellent book on the value of religions and taught us many things. But he wanted to be the prophet of a new religion. Instead of studying the effects of historical or existing forms of worship, he wanted to create a new one—an entirely different matter. So he gives just another illustration of the harm done to science by the mania for practical applications.

289. Spencer, on the other hand, after admitting, even too sweepingly, the influence of non-logical actions, eliminates them altogether by the general procedure described in § 261. Says he: "Our postulate must be that primitive ideas are natural, and, under the conditions in which they occur, rational."¹ Driven out by the door, logic here climbs back through the window. "In early life we have been taught that human nature is everywhere the same. . . . This error we must replace by the truth that the laws of thought are everywhere the same; and that, given the data as known to him, the primitive man's inference is the reasonable inference" (§§ 701, 711).

290. In assuming any such thing, Spencer puts himself in the wrong in his controversy with Comte. If human beings always draw logical inferences from the data they have before them, and if they act in accordance with such inferences, then we are left with nothing but logical conduct, and it is ideas that "govern or upset the world." There is no room left for those sentiments to which Spencer was disposed to attribute that capacity; there is no way for them to crowd into a ready-made aggregate composed of experimental facts, however badly observed, and of logical inferences derived from such facts.

291. The principle advanced by Spencer makes sociology very easy, especially if it be combined with two other Spencerian principles: unitary evolution, and the identity, or quasi-identity, of the savages of our time with primitive man (§§ 728, 731). Accounts by travellers, more or less accurate and more or less soundly interpreted, give us, Spencer thinks, the data that primitive man had at his disposal; and where such accounts fail, we fill in the gaps with our imagination, which, when it cannot get the real, takes the plausible. That gives us all we need for a sociology, for we have only to determine the logical implications of the data at hand, without wasting too much time on long and difficult historical researches.

292. In just that way Spencer sets about discovering the origin and evolution of religion. His primitive man is like a modern scientist working in a laboratory to frame a theory. Primitive man of course has very imperfect materials at his disposal. That is why, despite his logical thinking, he can reach only imperfect conclusions. All the same he gets some philosophical notions that are not a little subtle. Spencer, Ibid., Vol. I, § 154, represents as a "primitive" idea the notion that "any property characterizing an aggregate inheres in all parts of it." If you are desirous of testing the validity of that theory you need only state the proposition to some moderately educated individual among your friends, and you will see at once that he will not have the remotest idea of what you are talking about. Yet Spencer, loc. cit., believes that your friend will go on and draw logical conclusions from something he does not understand: "The soul, present in the body of the dead man preserved entire, is also present in preserved parts of his body. Hence the faith in relics." Surely Spencer could never have discussed that subject with some good Catholic peasant woman on the Continent. The argument he maps out might possibly lead a philosopher enamoured of logic to believe in relics, but it has nothing whatever to do with popular beliefs in relics.

293. So Spencer's procedure has points of similarity with Comte's procedure. In general terms, one might state the situation in this fashion: we have two things, P and Q (Figure 8), that have to be considered in determining the social order R. We begin by asserting that Q alone determines that order; then we show that P determines Q. So Q is eliminated, and P alone determines the social order.

294. If Q designates "ideas" and P "sentiments," we get, roughly, the evolution of Comte's theories. If Q designates "sentiments" and P "ideas," we get, roughly, the evolution of Spencer's theories.

295. That is confirmed by the remarks of John Stuart Mill on the controversy between Comte and Spencer. Says he:¹ "It will not be found, on a fair examination of what M. Comte has written, that he has overlooked any of the truth that there is in Mr. Spencer's theory. He would not indeed have said (what Mr. Spencer apparently wishes us to say) that the effects which can be historically traced, for example, to religion, were not produced by the belief in God, but by reverence and fear of Him. He would have said that the reverence and fear presuppose the belief: that a God must be believed in before he can be feared or reverenced." That is the very procedure in question! P is the belief in God; Q, sentiments of fear and reverence; P produces Q, and so becomes the cause determining conduct!

296. To a perfect logician like Mill it seems absurd that anyone could experience fear unless the feeling be logically inferred from a subject capable of inspiring fear. He should have remembered the verse of Statius,

"Primus in orbe deos fecit timor,"¹

297. In other connexions, Mill is perfectly well aware of the social importance of non-logical actions. But he at once withdraws the concession, in part at least, and instead of going on with what is, turns to speculations as to what ought to be. That is the general procedure; and many writers resort to it to be rid of non-logical conduct.

298. In his book On Liberty, p. 16, Mill writes, for example: "Men's opinions . . . on what is laudable or blameable, are affected by all the multifarious causes which influence their wishes in regard to the conduct of others, and which are as numerous as those which determine their wishes on any other subject: sometimes their reason—at other times their prejudices or superstitions: often their social affections, not seldom their antisocial ones, their envy or jealousy, their arrogance or contemptuousness: but most commonly, their desires or fears for themselves—their legitimate or illegitimate self-interest. Wherever there is an ascendant class, a large portion of the morality of the country emanates from its class interests, and its feelings of class superiority."

All that, with a few reservations, is well said and approximately pictures the facts.¹ Mill might have gone on in that direction, and inquired, since he was dealing with liberty, into the relations of liberty to the motives he assigns to human conduct. In that event, he might have made a discovery: he might have seen that he was involved in a contradiction in trying with all his might to transfer political power to "the greatest number," while at the same time defending a "liberty" that was incompatible with the prejudices, sentiments, and interests of said "greatest number." That discovery would then have enabled him to make a prophecy—one of the fundamental functions of science; namely, to foresee that liberty, as he conceived it, was progressively to decline, as being contrary to the motives that he had established as determinants of the aspirations of the class which was about to become the ruling class.

299. But Mill thought less of things as they are than of things as they ought to be. He says, Ibid., p. 22: "He [a man] cannot rightfully be compelled to do or forbear because it will be better for him to do so, because it will make him happier, because, in the opinions of others, to do so would be wise, or even right. These are good reasons for remonstrating with him, or reasoning with him, or persuading him, or entreating him, but not for compelling him, or visiting him with any evil in case he do otherwise."¹

That may be a "good justice," but it is not the justice handed out to us by our masters, who each year favour us with new laws to prevent our doing the very things that Mill says people should be allowed to do. His preaching, therefore, has been altogether without effect.

300. In certain writers the part played by non-logical actions is suppressed altogether, or rather, is regarded merely as the exceptional part, the "bad" part. Logic alone is a means to human progress. It is synonymous with "good," just as all that is not logical is synonymous with "evil." But let us not be led astray by the word "logic." Belief in logic has nothing to do with logico-experimental science; and the worship of Reason may stand on a par with any other religious cult, fetishism not excepted.

301. Condorcet expresses himself as follows:¹ "So a general knowledge of the natural rights of man; the opinion, even, that such rights are inalienable and unprescribable; a prayer voiced aloud for liberty of thought and press, for freedom of commerce and industry, for succour of the people . . . indifference to all religions —classified, at last, where they belong with superstitions and political devices [The good soul fails to notice that his worship of Progress is itself a religion!]—hatred and hypocrisy and fanaticism; contempt for prejudices; zeal for the propagation of enlightenment—all became the common avowal, the distinguishing mark, of anyone who was neither a Machiavellian nor a fool." Preaching religious toleration, Condorcet is not aware that he is betraying an intolerance of his own when he treats dissenters from his religion of Progress the way the orthodox have always treated heretics. It is true that he considers himself right and his adversaries wrong, because his own religion is good and theirs bad; but that, inverting terms, is exactly what they say too.

302. Maxims from Condorcet and other writers of his time are still quoted by humanitarian fanatics today. Condorcet continues, p. 292: "All errors in politics and morals are based on philosophical errors, which are in turn connected with errors in physic. There is no religious system, no supernatural extravagance, that is not grounded on ignorance of the laws of nature." But he himself gives proof of just such ignorance when he tries to have us swallow absurdities like the following, p. 345: "What vicious practice is there, what custom contrary to good faith, nay, what crime, that cannot be shown to have its cause and origin in the laws, the institutions, the prejudices, of the country in which that practice, that custom, is observed, that crime committed?" And he concludes finally, p. 346, that "nature links truth, happiness, and virtue with chain unsunderable."

303. Similar ideas are common among the French philosophes of the later eighteenth century. In their eyes every blessing doth from "reason" flow, every ill from "superstition." Holbach sees the source of all human woe in error;¹ and that belief has endured as one of the dogmas of the humanitarian religion, holiest of holies, of which our present-day "intellectuals" form the priesthood.²

304. All these people fail to notice that the worship of "Reason," "Truth," "Progress," and other similar entities is, like all cults, to be classed with non-logical actions. It was born, it has flourished, and it continues to prosper, for the purpose of combating other cults, just as in Graeco-Roman society the oriental cults arose out of opposition to the polytheistic cult. At that time one same current of non-logical conduct found its multiple expression in the taurobolium, the criobolium, the cult of Mithras, the growing importance of mysteries, Neo-Platonism, mysticism, and finally Christianity, which was to triumph over rival cults, none the less borrowing many things from them. So, toward the end of the eighteenth century and the beginning of the nineteenth, one same current of non-logical conduct finds its expression in the theism of the philosophes, the sentimental vagaries of Rousseau, the cult of "Reason" and the "Supreme Being," the love of the First Republic for the number 10, theophilanthropy (of which the "positivist" religion of Comte is merely an offshoot), the religion of Saint-Simon, the religion of pacifism, and other religions that still survive to our times.

These considerations belong to a much more comprehensive order, properly relating to the subjective aspect of theories indicated in § 13. In general, in other words, we have to ask ourselves why and how individuals come to evolve and accept certain theories. And, in particular, now that we have identified one such purpose—the purpose of giving logical status to conduct that does not possess it—we have to ask by what means and devices that purpose is achieved. From the objective standpoint, the error in the arguments just noted lies in their giving an a priori answer to the questions stated in § 14, and in maintaining that a theory needs simply to be in accord with the facts to be advantageous to society. That error is usually supplemented by the further error of considering facts not as they stand in reality but as they are pictured by the exhilarated imagination of the enthusiast.

305. Our induction so far has shown from some few particular cases the prevalence of a tendency to evade consideration of non-logical actions, which nevertheless force themselves upon the attention of anyone undertaking to discuss human societies; and also the no mean importance of that tendency. Now we must look into it specially and in general terms.¹

306. So let us now examine the various devices by which non-logical actions are eliminated so that only logical actions are left: and suppose we begin as usual by classifying the objects we are trying to understand.

Class A¹

The principles² underlying non-logical actions are held to be devoid of any objective reality (§§ 307-18).

Genera

Genus I. They are disregarded entirely (§§ 307-08).

Genus II. They are regarded as absurd prejudices (§§ 309-11)

Genus III. They are regarded as tricks used by some individuals to deceive others (§§ 312-18)

Class B

The principles underlying non-logical actions are credited with now more, now less, objective reality (§§ 319-51)

Genera and Subgenera

Genus I. The principles are taken as completely and directly real (§§ 319-38)

Iα. Precepts with sanctions in part imaginary (§§ 321-33)

Iβ. Simple interposition of a personal god or a personified abstraction (§§ 332-33)

Iγ. The same interposition supplemented by legends and logical inferences (§ 334)

Iδ. Some metaphysical entity is taken as real (§§ 335-36)

Iε. What is real is an implicit accord between the principles and certain sentiments (§§ 337-38)

Genus II. The principles of non-logical conduct are not taken as completely or directly real. Indirectly, the reality is found in certain facts that are said to be inaccurately observed or imperfectly understood (§§ 339-50)

IIα. It is assumed that human beings make imperfect observations, and derive inferences from them logically (§§ 340-46)

IIβ. A myth is taken as the reflection of some historical reality that is concealed in one way or another, or else as a mere imitation of some other myth (§§ 347-49)

IIγ. A myth is made up of two parts: a historical fact and an imaginary adjunct (§ 350)

Genus III. The principles of non-logical actions are mere allegories (§§ 351-52)

Class C

It is assumed that non-logical actions have no effect on "progress," or else are obstructive to it. Hence they are to be eliminated in any study designed solely to promote "progress" (§§ 353-56).

307. Let us examine these various categories one by one.

Device A-I: Non-logical actions are disregarded. Non-logical actions can be disregarded entirely as having no place in the realm of reality. That is the position of Plato's Socrates in the matter of the national religions of Greece.¹ He is asked what he thinks of the ravishing of Orithyia, daughter of Erechtheus, by Boreas. He begins by rejecting the logical interpretation that tries to see a historical fact in the myth (IIγ). Then he opines that such inquiries are as finespun as they are profitless, and falls back on the popular belief. On common belief the oracle at Delphi also relied when it prescribed that the best way to honour the gods was for each to follow the customs of his own city.² Certainly the oracle in no wise meant by that that such customs corresponded to things that were not real; yet actually they might as well have, since they were held to be entirely exempt from the verification to which real things are considered subject. That method often amounts to viewing beliefs as non-logical actions to be taken for what they are without any attempt to explain them—the problem being merely to discover the relationship in which they stand towards other social facts. That, overtly or tacitly, is the attitude of many statesmen.

308. So, in Cicero's De natura deorum, the pontifex Cotta distinguishes the statesman from the philosopher. As pontifex he protests that he will ever defend the beliefs, the worship, the ceremonies, the religion, of the forefathers, and that no argument, be it of scholar or dunce, will ever budge him from that position. He is persuaded that Romulus and Numa founded Rome, the one with his auspices, the other with his religion. "That, Balbus, is what I think, as Cotta and as pontifex. It is now for me to know what you think. From you, a philosopher, I have a right to expect some reason for your beliefs. The beliefs I get from our forefathers I must accept quite apart from any proof."¹ In that it is obvious that as pontifex Cotta deliberately steps aside from the realm of logical reality, which implies a belief either that traditional Roman beliefs have no basis in fact or else that they are to be classed with non-logical actions.²

309. Device A-II: The principles of non-logical actions are regarded as absurd prejudices. One may consider merely the forms of non-logical actions and finding them irrational, judge them absurd prejudices, at the most deserving of attention from a pathological standpoint as veritable maladies of the human race. That has been the attitude of not a few writers in dealing with legal and political formalities. It is the attitude especially of writers on religion and most of all of writers on forms of worship. It is also the attitude of our contemporary anti-clericals with regard to the Christian religion—and it betrays great ignorance on the part of those bigots, along with a narrow-mindedness that incapacitates them for ever understanding social phenomena.

We have already seen specimens of this type of reasoning in the works of Condorcet (§§ 301-02) and Holbach (§§ 296², 303). A more diluted type is observable in disquisitions purporting to make this or that religion "more scientific" (§ 16²), on the assumption that a religion which is not scientific is either absurd or reprehensible. So in earlier times there were efforts to remove by subtle interpretation such elements in the legends and cults of the pagan gods as were considered non-logical. It was the procedure of the Protestants during the Reformation, while the liberal Protestants of our day are repeating the same exploits, appealing to their pseudo-science. So also for the Modernists in their criticism of Catholicism, and for our Radical Socialists in their demeanour towards Marxism.

310. If one regards certain non-logical actions as absurd, one may centre chiefly on their ridiculous aspects; and that is often an effective weapon for combating a faith. Frequent use of it was made against established religions from the day of Lucian down to the day of Voltaire. In an article replete with historical blunders, Voltaire says of the religion of Rome: "I am imagining that after conquering Egypt Caesar sends an embassy to China, with the idea of stimulating the foreign trade of the Roman Empire. . . . The Emperor Iventi, first of that name, is reigning at the time. . . . After receiving Caesar's ambassadors with typical Chinese courtesy, he secretly inquires through his interpreters as to the civilization, customs, and religion of these Romans. . . . He learns that the Roman People supports at great expense a college of priests, who can tell you exactly the right time for embarking on a voyage and the very best place for fighting a battle by inspecting the liver of an ox or the appetite with which chickens eat their barley. That sacred science was brought to the Romans long, long before by a little god named Tages, who was unearthed somewhere in Tuscany. The Roman people worship just one god whom they always call 'Highest and Best.' All the same, they have built a temple to a harlot named Flora; and most Roman housewives have little household gods in their homes, five or six inches high. One of the little divinities is the goddess Nipples, another the god Bottom. . . . The Emperor has his laugh. The courts at Nanking at first conclude, as he does, that the Roman ambassadors are either lunatics or impostors . . . but the Emperor, being as just as he is courteous, holds private converse with the ambassadors. . . . They confess to him that the College of Augurs dates from early ages of Roman barbarism; that an institution so ridiculous has been allowed to survive only because it became endeared to the people in the course of long ages; that all respectable people make fun of the augurs; that Caesar never consults them; that according to a very great man by the name of Cato no augur is ever able to speak to a colleague without a laugh; and finally that Cicero, the greatest orator and best philosopher of Rome, has just published against the augurs a little essay, On Divination, in which he hands over to everlasting ridicule all auspices, all prophecy, and all the fortune-telling of which humanity is enamoured. The Emperor of China is curious to read Cicero's essay. His interpreters translate it. He admires the book and the Roman Republic."¹

311. In dealing with writings of this kind, we must be careful not to fall into the very error we are here considering, with reference to non-logical actions.¹ The intrinsic value of such satires may be zero when viewed from the experimental standpoint, whereas their polemical value may be great. Those two things we must always keep distinct. Moreover they may have a certain intrinsic value: a group of non-logical actions taken as a whole may be useful for attaining a given purpose without absolutely all of them, taken individually, being useful to that purpose. Certain ridiculous actions may be eliminated from such a group without impairing its effectiveness. However, in so reasoning we must beware of falling into the fallacy of the man who said he could lose all his hair without becoming bald because he could lose any particular hair without suffering that catastrophe.

312. Device A-III: Non-logical actions as tricks for deceit. After establishing, as in the two cases above, that certain actions are not logical, but still resolved to have them such in the feeling that every human act should be born of logic, a writer may go on and say that an institution involving non-logical conduct is an invention of this or that individual or group that is designed to procure some personal advantage, or some advantage to state, society, or humanity at large. So actions intrinsically non-logical are transformed into actions that are logical from the standpoint of the end in view.

To adopt this procedure as regards actions deemed beneficial to society is to depart from the extreme case noted in § 14, where it is maintained that only theories which accord with facts (logico-experimental theories) can be beneficial to society. It is here recognized that there are theories which are not logico-experimental, but which are nevertheless beneficial to society. All the same, the writer cannot make up his mind to admit that such theories derive spontaneously from non-logical impulse. No, all conduct has to be logical. Therefore such theories too are products of logical actions. These actions cannot originate in the sources of the theories, since it has been recognized that the theories have no experimental basis; but they may envisage the same purposes as the theories, which experience shows are beneficial to society. So we get the following solution: "Theories not in accord with the facts may be beneficial to society and are therefore logically invented to that end."¹

313. The notion that non-logical actions have been logically devised to attain certain purposes has been held by many many writers. Even Polybius, a historian of great sagacity, speaks of the religion of the Romans as originating in deliberate artifice.¹ Yet he himself recognized that the Romans succeeded in creating their commonwealth not by reasoned choices but by allowing themselves to be guided by circumstances as they arose.²

314. We may take Montesquieu's view of Roman religion as the type of the interpretation here in question.¹ "Neither fear nor piety established religion among the Romans, but the same necessity that compels all societies to have religions. . . . I note this difference, however, between Roman legislators and the lawgivers of other peoples, that the Romans created religion for the State, the others the State for religion. Romulus, Tatius, and Numa made the gods servants of statesmanship; and the cult and the ceremonies that they instituted were found to be so wise that when the kings were expelled the yoke of religion was the only one which that people dared not throw off in its frenzy for liberty. In establishing religion, Roman law-makers were not at all thinking of reforming morals or proclaiming moral principles. . . . They had at first only a general view, to inspire a people that feared nothing with fear of the gods, and to use that fear to lead it whithersoever they pleased. . . . It was in truth going pretty far to stake the safety of the State on the sacred appetite of a chicken and the disposition of the entrails in a sacrificial animal; but the founders of those ceremonies were well aware of their strong and weak points, and it was not without good reasons that they sinned against reason itself. Had that form of worship been more rational, the educated as well as the plain man would have been deceived by it; and so all the advantage to be expected from it would have been lost."

315. It is curious that Voltaire and Montesquieu followed opposite though equally mistaken lines, and that neither of them thought of a spontaneous development of non-logical conduct.

316. The variety of interpretation here in question sometimes contains an element of truth, not as regards the origin of non-logical actions, but as regards the purposes to which they may be turned once they have become customary. Then it is natural enough that the shrewd should use them for their own ends just as they use any other force in society. The error lies in assuming that such forces have been invented by design (§ 312). An example from our own time may bring out the point more clearly. There are plenty of rogues, surely, who make their profit out of spiritualism; but it would be absurd to imagine that spiritualism originated as a mere scheme of rogues.

317. Van Dale, in his treatise De Oraculis, saw nothing but artifice in the pagan oracles. That notion belongs with this group of interpretations. Eusebius wavers between it and the view that oracles were the work of devils.¹ Such mixtures of interpretations are common. We shall come back to them.

318. Likewise with this variety are to be classed interpretations that regard non-logical actions as consequences of an external or exoteric doctrine serving to conceal an internal or esoteric doctrine. That would make actions which are non-logical in appearance logical in reality. Consider a passage in Galileo's Dialogue of the Greater Systems (Salviati speaking):¹ "That the Pythagoreans held the science of numbers in very high esteem . . . I am well aware, nor would I be loath to concur in that judgment. But that the mysteries in view of which Pythagoras and his sect held the science of numbers in such great veneration are the absurdities commonly current in books and conversation, I can in no way agree. On the contrary, they did not care to have their wonders exposed to the ridicule and disparagement of the common herd. So they damned as sacrilegious any publication of the more recondite properties of the numbers and incommensurable and irrational quantities with which they dealt, and they preached that anyone disclosing such things would suffer torment in the world to come. I think that some of them, to throw a sop to the vulgar and be free of prying importunity, represented their numeral mysteries as the same childish idiocies that later on spread generally abroad. It was a shrewd and cunning device on their part, like the trick of that sagacious young man who escaped the prying of his mother (or his curious wife—I forget which), who was pressing him to confide the secrets of the Senate, by making up a story wherewith she and other prattling females proceeded to make fools of themselves, to the great amusement of the Sentaors."

That the Pythagoreans sometimes misrepresented their own doctrines seems certain; but it is not at all apparent that that was the case with their ideas on perfect numbers. On that point Galileo is mistaken (§§ 960 f.).

319. Device B-I: The principles are taken as completely and directly real.¹ This variety is exemplified by non-logical actions of a religious character on the part of unquestioning believers. Such actions differ little if at all from logical actions. If a person is convinced that to be sure of a good voyage he must sacrifice to Poseidon and sail in a ship that does not leak, he will perform the sacrifice and caulk his seams in exactly the same spirit.

320. Curiously enough, such doctrines come closer than any others to a scientific status. They differ from the scientific, in fact, only by an appendage that asserts the reality of an imaginary principle; whereas many other doctrines, in addition to possessing the same appendage, further differ from scientific doctrines by inferences that are either fantastic or devoid of all exactness.

321. Device B-Iα: Precept plus sanction. This variety is obtained by appending some adjunct or other to the simple sanctionless precept—to the taboo (cf. § 154).¹

322. Reinach writes:¹ "A taboo is an interdiction; an object that is taboo, or tabooed, is a forbidden object. The interdiction may forbid corporal contact or visual contact; it may also exempt the object from the peculiar kind of violation involved in pronouncing its name. . . . Similar interdictions are observable in Greece and Rome, and among many other peoples, where generally it is explained that knowledge of a name enables a person to 'evoke' with evil intent the 'power' that the name designates. That explanation may have been valid at certain periods; but it does not represent the primitive state of mind. Originally it was the sanctity of the name itself that was dreaded, on the same grounds as contact with a tabooed object."

Reinach is right in regarding as an appendage the notion that knowledge of the name of an object gives a person power over it; but the notion of sanctity is likewise an appendage. Indeed, probably few of the individuals observing a taboo would know what was meant by an abstraction such as "sanctity." For them the taboo is just a non-logical action, just an aversion to touching, looking at, naming, the thing tabooed. Later on an effort is made to explain or justify the aversion; and then the mysterious power of which Reinach speaks (or perhaps his own notion of sanctity) is invented.

Reinach continues: "The notion of the taboo is narrower still than the notion of interdiction. The characteristic difference is that the taboo never gives a reason." That is excellent! The non-logical action has just that trait. But for that very reason Reinach should not, in a particular case, provide the taboo with a reason in some consideration of sanctity. He goes on: "The prohibition is merely stated, taking the cause for granted—it is, in fact, nothing but the taboo itself, that is to say, the assertion of a mortal peril." But in saying that he is withdrawing his concession and trying to edge back into the domain of logic. No "cause" is taken for granted! The taboo lies in a pure and absolute repugnance to doing a certain thing. To get something similar from our own world: There is the sentimental person who could never be induced to cut off a chicken's head. There is no "cause" for the aversion; it is just an aversion, and it is strong enough to keep the person from cutting off a chicken's head! It is not apparent either why Reinach would have it that the penalty for violating a taboo is always a mortal peril. He himself gives examples to the contrary. Going on, he returns to the domain of non-logical actions, well observing that "the taboos that have come down into contemporary cultures are often stated with supporting reasons. But such reasons have been excogitated in times relatively recent [One could not say better.] and bear the stamp of modern ideas. For example, people will say, 'Speak softly in a chamber of death [A taboo that gives no evidence of having a "mortal peril" for a sanction.] out of respect for the dead.' The primitive taboo lay in avoiding not only contact with a corpse, but its very proximity. [Still no evidence of any mortal peril.] Nevertheless even today, in educating children taboos are imparted without stated reasons, or else with some mere specification of the general character of the interdiction: 'Do not take off your coat in company, for that is not nice.' In his Works and Days, v. 727, Hesiod interdicts passing water with one's face towards the sun, but he gives no reasons for the prohibition. [A pure non-logical action.] Most taboos relating to decorum have come down across the centuries without justifications" [and with no threats of "mortal peril"].²

323. With taboos may profitably be classed other things of the kind where logical interpretation is reduced to a minimum. William Marsden says of the Mohammedans of Sumatra:¹ "Many who profess to follow it [Mohammedanism] give themselves not the least concern about its injunctions, or even know what they require. A Malay at Marina upbraided a countryman, with the total ignorance of religion, his nation laboured under, 'You pay a veneration to the tombs of your ancestors: what foundation have you for supposing that your dead ancestors can lend you assistance?' 'It may be true,' answered the other; 'but what foundation have you, for expecting assistance from Allah and Mahomet?' 'Are you not aware,' replied the Malay, 'that it is written in a Book? Have you not heard of the Koraan?' The native of the Passumah, with conscious inferiority, submitted to the force of this argument."² That is a seed which will sprout and yield an abundant harvest of logical interpretations, some of which we shall find in the devices hereafter following.

324. Something like the taboo is the precept (§§ 154, 1480 f.). It may be given without sanction, "Do so and so," and in that form it is a plain non-logical action. In the injunction, "You ought to do so and so," there is a slight, sometimes a very slight, trace of explanation. It lurks in the term "ought," which suggests the mysterious entity Duty. That is often supplemented by a sanction real or imaginary, and then we get actions that are either actually logical or else are merely made to appear so. Only a certain number of precepts, therefore, can be properly grouped with the things we are classifying here.

325. In general, precepts may be distinguished as follows:

a. Pure precept, without stated reasons, and without proof. The proposition is not elliptical. No proof is given, either because no proof exists or because none is asked for. That, therefore, is the pure non-logical action. But human beings have such a passion for logical explanations that they usually stick one or two on, no matter how silly. "Do that!" is a precept. If it be asked, "Why should I do that?" the answer is, let us say, "Because  . . . !" or, "Because it is customary." The logical appendage is of little value, except where violation of custom implies some penalty—but in that case the penalty, not the custom, carries the logical force.

326. b. The demonstration is elliptical. The proof, valid or not, is available. It has not been mentioned, but it may be. The proposition is a precept only in appearance. The terms "ought," "must," and the like may be suppressed, and the precept reduced to an experimental or pseudo-experimental theorem, the consequence deriving from the act without any interposition from without. This type of precept runs, "To get A, you must do B,"; or, negatively, "To avoid A, you must refrain from doing B." The first proposition can be stated thus: "When B is done, A results." Similarly for the second.

327. If both A and B are real things and if the nexus between them is actually logico-experimental, we get scientific propositions. They have nothing to do with the things we are trying to classify here. If the nexus is not logico-experimental, they are pseudo-scientific propositions, and a certain number of them are used to logicalize non-logical actions. For instance, if A stands for a safe voyage and B for sacrifices to Poseidon, the nexus is imaginary, and the non-logical action B is justified by the nexus that connects it with A. But if A stands for a safe voyage and B for defective ship-building, we get just an erroneous scientific proposition. A mistake in engineering is not a non-logical action.

328. If A and B are both imaginary, we are wholly outside the experimental field, and we need not consider such propositions. If A is imaginary and B real, we get non-logical actions, B, justified by the pretext, A.

329. c. The proposition is really a precept, but a real sanction enforced by an extraneous and real cause is appended to it. That gives a logical action: the thing is done to escape the sanction.

330. d. The proposition is a precept, but the sanction is imaginary, or enforcible only by an imaginary power. We get a non-logical action justified by the sanction.¹

331. The terms of ordinary speech rarely have sharply defined meanings. The term "sanction" may be used more or less loosely. Here we have taken it in the strict sense. Broadly speaking, one might say that a sanction is always present. In the case of a scientific proposition the sanction might be the pleasure of reasoning soundly or the pain of reasoning amiss. But to go into such niceties would be just a waste of time.

332. Device B-Iβ: Introduction of a divinity or of personified abstractions. A very simple elaboration of the taboo, or pure precept, is involved in the introduction of a personal god, or of personifications such as Nature, by will of which non-logical actions are required of human beings and are therefore logicalized. How the requirement arises is often left dark. "A god (or Nature) wills that so and so be done." "And if it is not done?" The question remains unanswered. But very often there is an answer; it is asserted that the god (or Nature) will punish violators of the precept. In such a case we get a sanctioned precept of the species d above.

333. When the Greeks said that "strangers and beggars come from Zeus,"¹ they were merely voicing their inclination to be hospitable to visitors, and Zeus was dragged in to give a logical colouring to the custom, by implying that the hospitality was offered either in reverence for Zeus, or to avoid the punishment that Zeus held in store for violators of the precept.

334. Device B-Iγ: Divinities plus legend and logical elaboration. Rare the case where such embellishments are not supplemented by multiple legends and logical elaborations; and through these new adjuncts we get mythologies and theologies that carry us farther and farther away from the concept of non-logical conduct. It may be worth while to caution that theologies at all complicated belong to restricted classes of people only. With them we depart from the field of popular interpretations and enter an intellectual or scholarly domain. To the variety in question here belong the interpretations of the Fathers of the Christian Church, such as the doctrine that the pagan gods were devils.

335. Device B-Iδ: Metaphysical entities taken as real. Here reality is ascribed not to a personal god or to a personification, but to a metaphysical abstraction. "The true," "the beautiful," "the good," "the honest," "virtue," "morality," "natural law," "humanity," "solidarity," "progress," or their opposite abstractions, enjoin or forbid certain actions, and the actions become logical consequences of the abstractions.¹

336. In interpretations of the B-Iβ variety, the personal god can inflict a punishment because he chooses to. In the case of "Nature" the punishment is an automatic consequence of the conduct. Those interpretations, therefore, are respectably logical. In the case of metaphysical abstractions, however, the logic is flimsy indeed. You tell a person, "You must do that because it is good," and he replies, "But I do not choose to do what is good." You are checkmated, for milord Good, estimable worthy that he may be, does not wield the thunderbolts that Zeus wields. So our latter-day Christians keep the God of the Old Testament but strip Him of all His weapons. There could be no trifling with the God of the Hebrews, who fiercely avenged transgressions of His laws, or with the God of St. Paul, who was no whit less quick to wrath. But, armed with the abstractions of their pseudo-science, with what can the neo-Christians threaten the unbeliever? Or what can they do for the believer to make his belief worth while? The answer is, "Nothing." The conduct they recommend is simply non-logical conduct. That does not mean that it may not be as beneficial to individual or society as any other, or even more so. It may or may not be. But in any event it is certain that it is not the logical inference from a principle, like the inference from the existence of a divine power and will that unbelievers will be punished and believers rewarded.¹

337. Device B-Iε: What is real is the accord between the principles and certain sentiments. This manner of envisaging facts is implicit rather than explicit. So for certain neo-Christians the reality of Jesus seems to come down to an accord between their conception of Him and certain sentiments they hold. They abandon the objective field, deny the divine nature of Christ, and seem not to care very much about His historical reality. They are satisfied with asserting that Christ is the most perfect type of humanity, which means that their notions of Christ happen to coincide with what, according to their sentiments, is the most perfect type of human being. Once on that road they finish by throwing all theology, all rites, overboard and end with the assertion that "religion is a manner of living."¹

338. Along that line they might seem to be approximating the concept of non-logical conduct; but they are still radically at variance with it, since they are thinking not of what is, but of what ought to be, and rob the "ought" of the subordinate character (§ 326) it might well have in the case of some few individuals, and give it an absolute status that altogether transcends the experimental field. Their theories, in a word, have no other purpose than to decorate non-logical impulse with a logical rouge.

339. B-II: The reality is no longer direct; that is to say, it is no longer held that there is a god, a personification, an abstraction, or the like, from which non-logical actions may be logically inferred. It is assumed that such actions have arisen spontaneously, by reasonings good or bad based on facts well or badly authenticated. The difference between this variety and the B-I group is a radical one; for whereas the B-I devices ascribed reality to entities foreign to the experimental field, the entities posited in this variety arise within the experimental field, and the only questions are whether they have actually been observed and whether the assumed consequences are real consequences. "Beggars come from Zeus" is an interpretation of the B-I variety. I create the entity Zeus, which I assume to be real, and from its existence I draw certain inferences. "Whoever is hospitable to beggars will be happy" is an interpretation of the B-II variety. I pretend that I have observed that people who have been hospitable to beggars have been happy, and I draw the inference that if they continue to be hospitable to beggars they will continue to be happy. I have not created any entity; I am using real facts, combining them as I see fit.

340. Device B-IIα: Observation imperfect, inferences logical. This method of reasoning aims to throw back upon the premises a logico-experimental insufficiency that cannot be disputed. We have certain assertions that are manifestly in contradiction with logico-experimental knowledge. We may assume that the contradiction arises because the reasoning which produces the conclusions is not logical, and we are thereby carried into the domain of non-logical conduct. Or else we may hold that the reasoning is logical, but that it starts with premises inconsistent with experimental knowledge and so leads to conclusions where the contradiction is likewise apparent. In that way we are able to remain within the field of logical conduct. Typical of this variety are the theories of Herbert Spencer (§§ 285, 289-95). The rôle ascribed to non-logical conduct is reduced to a minimum and may even be eliminated. Underlying certain phenomena are certain observations of fact. It is assumed that from such alleged observations human beings have drawn inferences, reasoning very much as any thinker would reason. So we get the doctrines of those human beings and the reasons for their conduct.

341. Concepts of this kind figure to a greater or lesser extent in almost all theories dealing with the "origins" of social phenomena such as "religion," "morality," "law," and the like. Writers are driven to admit the existence of non-logical actions but are careful to push them back into the past as far as they can.

342. There may be some truth in such theories in so far as they call attention to certain simple types of complex phenomena. They go astray in trying to derive the complex phenomenon from the simple type, and still farther astray when it is assumed that that process is logical.

343. Ignoring for the moment the complex character of social phenomena, let us assume that certain phenomena P, observable at the present day, have an actual origin A (Figure 9). If the development took place along a continuous line ABCDP, it would be possible, in a sense, to take one of the intermediate phenomena B, C . . . as the origin, or cause, of P. If, for instance, going as far back as our historical knowledge permits, we found a thing B of the same nature as P, though much simpler, we should not go too far wrong in regarding it as the origin, or cause, of P.

344. Unfortunately the assumption of development along a continuous line does not at all conform with the facts as regards social phenomena, or even as regards not a few biological phenomena. The development, rather, seems to take place along a line with many branches (Figure 10), even still ignoring the complex character of social phenomena, which hardly permits us to dissociate the social phenomenon P from other social phenomena (§ 513). Facts B, C, D . . . (Figure 10) are no longer located along a straight continuous line, but stand at the extremities or intersections of branch lines; and we cannot, even as a hypothesis very remotely approximative to the facts, assume that C, for example, or E, or any other similar fact observable in the past, is the origin, the cause, of P, observable in the present.

345. To take a concrete example: Reinach sees in taboos the origin of religion. In so doing, he seems to take the position pictured in Figure 9. B standing for the taboos, P for present-day religions. But even assuming that religion is unconnected with other social phenomena, the situation is actually as represented in Figure 10, and the taboos B would be the extremity of a by-path. Taboos cannot be taken as the origin of religion. They may be regarded as simple types of phenomena, of which the religions C, Q, P are complex types. That is all the truth there is in the theories of Reinach, a fairly important truth, for that matter, since it emphasizes the part played by non-logical actions in religious phenomena.

346. Studies in origins in social matters often proceed very much after the manner of old-fashioned etymology.¹ The intermediate steps C, D . . . (Figure 9) are assumed or guessed at, in getting from B to P; and the temptation is to ask how things ought to have gone rather than how they actually went. Investigations, in such a case, lie outside the domain of experimental reality. Yet, historically speaking, they have not been altogether wasted: for they have served to open a breach in the ethical and a priori theories that have been explaining P by imaginary principles. That task accomplished, it is now time for them to give way to purely experimental theories.

347. Device B-IIβ: Myths have a historical basis or else are imitations of other myths. Origins and evolution being discarded, it is assumed that every myth is the deformed reflection of something real. Of this variety were the euhemeristic theories, so called, as to the origin of the pagan gods (§§ 682-708). Nothing is more certain than that there have been cases where human beings have been deified. The euhemeristic error lies, first of all, in generalizing a particular fact, and then in confusing the point B in Figure 9 with the point B in Figure 10, in assuming, that is, that because one fact precedes another fact in time, it is the origin of it. The theories of Palaephatus (§ 661) also belong to this variety.

348. In general, interpretations of this kind are very easy to work out. One arbitrarily changes in a myth anything that needs to be changed to produce a picture that is real. Take, for example, Astolfo's hippogriff in the Orlando furioso of Ariosto. The wingèd horse can be made a real horse by interpreting the story in the sense that the hippogrirt was some very swift horse that was therefore spoken of as having wings. Dante sees Francesca and her brother-in-law lashed by "the hellish hurricane." The hurricane can be interpreted as a symbol of the carnal passion that smites the two lovers like a hurricane. In such a procedure not the slightest difficulty will ever be encountered (§ 661).

349. With this variety we may class theories that explain the non-logical actions observable in a given society as imitations of non-logical actions prevalent in other societies. To tell the truth, not all non-logical actions are eliminated by this device; they are merely reduced in number, several of them being taken as duplicates of one.¹

350. Device B-IIγ: Myths taken as historical fact plus a fictional appendage. In this variety we come a little closer to reality. In every myth the legend is assumed to have a nucleus of historical fact covered over by an alluvium of fiction. One removes the accretion, and finds the nucleus of fact underneath. Many books have been written from that point of view. Not so long since all the legends that have come down from Graeco-Roman antiquity were treated in that way.¹

Our variety B-IIβ, above, is often the present variety, B-IIγ, carried to the extreme. There may be something historical in a myth, a something more or less extensive. As it is reduced to a minimum and finally disappears, we get the B-IIβ variety.

351. Device B-III: The principles underlying non-logical actions are allegories. The actions, it is held, are in reality logical. They seem to be non-logical only because the allegories are taken literally. A further assumption locates the source of such errors in language by an allegorical interpretation. Max Müller writes:¹ "There are many myths in Hesiod, of late origin, where we have only to replace a full verb by an auxiliary, in order to change mythical into logical language. Hesiod [Theogonia, vv. 211-12 (White, pp. 94-95)], calls Nyx (Night) the mother of Moros (Fate), and the dark Kêr (Destruction), of Thanatos (Death), Hypnos (Sleep) and the tribe of the Oneiroi (Dreams). . . . Now let us use our modern expressions, such as: 'the stars are seen as night approaches,' 'we sleep,' 'we dream,' 'we die,' 'we run danger during the night' . . . and we have translated the language of Hesiod . . . into modern forms of thought and speech."

352. On that basis all myths would be charades. It seems incredible that a theory so manifestly absurd could have gained such wide acceptance. Müller's disciples did even worse than their master, and the solar myth became a convenient and universal explanation for every conceivable legend.

353. Class C. In this class, really, non-logical actions are not interpreted in such a way as to make them logical. They are eliminated, so that only logical actions are left. That serves just as well to reduce all conduct to logic. Such opinions are widely current in our time, and are an article of faith with a great many people who worship a powerful divinity known to them as "Science." Not a few humanitarians are of the same tribe.

354. Other people reason more soundly; and after noting a thing that is true enough—that science has contributed greatly to the advance of civilization—they go farther still and try to show that nothing that is not science can be useful. As the type of such theories one might quote the celebrated argument of Buckle:¹ "It is evident, that if we look at mankind in the aggregate, their moral and intellectual conduct is regulated by the moral and intellectual notions prevalent in their own time. . . . Now, it requires but a superficial acquaintance with history to be aware that this standard is constantly changing. . . . This extreme mutability in the ordinary standard of human actions shows that the conditions on which the standard depends must themselves be very mutable; and these conditions, whatever they may be, are evidently the originators of the moral and intellectual conduct of the great average of mankind.

"Here, then, we have a basis on which we can safely proceed. We know that the main cause of human actions is extremely variable; we have only, therefore, to apply this test to any set of circumstances which are supposed to be the cause, and if we find that such circumstances are not very variable, we must infer that they are not the cause we are attempting to discover.

"Applying this test to moral motives, or to the dictates of what is called moral instinct, we shall at once see how extremely small is the influence those motives have exercised over the progress of civilization. For there is, unquestionably, nothing to be found in the world which has undergone so little change as those great dogmas of which moral systems are composed. . . .

"But, if we contrast this stationary aspect of moral truths with the progressive aspect of intellectual truths, the difference is indeed startling.² All the great moral systems which have exercised much influence have been fundamentally the same; all the great intellectual systems have been fundamentally different. . . . Since civilization is the product of moral and intellectual agencies, and since that product is constantly changing, it evidently cannot be regulated by the stationary agent; because when surrounding circumstances are unchanged, a stationary agent can only produce a stationary effect. The only other agent is the intellectual one; and that this is the real mover may be proved."

355. Buckle's reasoning is sound provided one add that all human conduct is logical and derives from moral and intellectual principles. But that proposition is false. In the first place, many very important actions are non-logical. Secondly, the things designated by the terms "moral principle" and "intellectual principle" are wanting in exactness: they cannot be taken as premises in a rigorous argument. Thirdly, Buckle's reasoning has the general defect of arguments by elimination in sociological matters—the enumeration is never complete.¹ He omits things of great importance. Theoretical principles of morality may be the same, and moral practices very different—for instance, the peoples who all preach the Christian ethics by no means all behave in the same way in practice.²

356. Buckle's argument reduces the practical rôle of moral theories to very small proportions, and in that it accords with the facts. But what it takes away from morals ought not be handed over to an "intellectual principle" (whatever that may be), but to the patrimony of non-logical actions, economic progress, improvements in communications, and the like. It may well be that something has to be assigned to scientific progress all the same, and therefore to the said "intellectual principle"; but there is a big difference between such indirect, non-logical influence, and a direct action by way of logical inference from a given principle.¹

357. We need carry our study of this special classification no farther. It has already shown that existing doctrines may be broken up into two different elements: certain sentiments, and inferences from those sentiments. It opens, in other words, a path that it may or may not be profitable to follow to the end. We shall see as we go on.

358. Many statesmen, many historians, recognize non-logical actions without giving them that name and without going to the trouble of finding their theory. Just a few examples taken here and there from the works of Bayle,¹ implicit in which are several theories of non-logical conduct—and it is indeed surprising to find in a writer who lived two centuries and more ago certain truths that are unappreciated even today. Bayle declares and repeats that "opinions are not the rule of conduct"; and that "man does not regulate his conduct by his opinions. . . . The Turks hold certain tenets of that doctrine of the Stoics [fatalism], and they carry the business of predestination to extreme lengths. Nevertheless they may be seen to flee danger as other men do, and they are far from charging in battle with the courage of the French, who do not believe in predestination." The existence and importance of non-logical conduct could not be recognized in plainer terms. Find a general form for this observation of particular fact, and we get the starting-point for a theory of non-logical conduct.

359. Bayle further observes, Ibid., § 139: "It cannot be said that people who fail to live according to the precepts of their religion do not believe in a God"; and he presses the point, Ibid., § 136: "Man does not act according to his principles. He may be as rational a creature as you like, but it is none the less true that he almost never acts according to his principles. [In other words his conduct is non-logical.] He has indeed the strength, in speculative matters, not to draw wrong conclusions; for in such reflections he sins rather in his readiness to accept false principles than in drawing mistaken conclusions from them. But it is quite another matter when good morals are in question. [A particular remark that is true in general.] In morals he almost never hits on false principles. Almost always the ideas of natural equity are present in his conscience. Nevertheless he is always deciding in favour of his uncontrolled desires. [The usual vague phraseology, but the substance accords with fact.] . . . The true principle of human conduct . . . is naught but temperament, the natural inclination to pleasure, the taste for certain things, the desire to please, the habits acquired in intercourse with friends, or some other disposition arising from the depths of human nature, whatever the country in which one is born [This contradicts the preceding and is to be deleted.] and whatever the knowledge that has been instilled in the mind."

That comes very close to the facts. If we tried to give greater precision to Bayle's language, and establish a stricter classification, would we not have a theory of non-logical actions—their great importance so becoming more and more apparent?

360. Bayle quotes with approval a passage from Nicolle: " 'When the time comes for human beings to pass from speculation to action, they do not follow consequences; and strange it is to see how the human mind can stop at certain speculative truths without going on to their logical consequences in practice, which seem so bound up with those truths as to be in no way separable from them.' "¹

361. Bayle soundly enough observes, Ibid., § 51, that "the pagan religion was satisfied with an external rite" (§ 174); but he went wrong in believing, Ibid., § 122, that it "had no influence on morals." He failed to perceive that ritual practices intensified sentiments (non-logical actions) and that such sentiments were in turn sources of morality.

362. He goes to some pains to prove that atheism is preferable to idolatry. To understand him aright we have to take account of the times in which he was living and the perils to which he was exposed. Just as in our time there are persons who give perpetual chase to "immoral" books, so in Bayle's time there were those who kept open season on books against Christianity. Unable to whip the horse, Bayle whips the saddle, and belabours idolatry with criticisms that apply just as well to all religions. At bottom his argument tends to show that since the majority of human actions are non-logical, forms of belief are of no great importance.

363. Montesquieu did not get that point, and his reply to what he calls "Bayle's paradox" is of little or no value. He is solving the problem by restating it when he says: "A prince who loves religion and fears it is a lion surrendering to the hand that caresses it, or to the voice that quiets it; the prince who fears religion and hates it is like the wild beast biting at the chains that keep it from attacking passers-by; the prince who has no religion at all is the terrible beast that never feels his freedom till he is rending and devouring."¹ Underlying all this declamation, which is mere fustian, is the proposition, evidently, that human beings act logically in accord with their beliefs. But that is the very thing Bayle denies; and proofs, not mere asseverations of the opposite, were required to refute him (§ 368).

364. Taking his stand on logical conduct, Montesquieu says that "even if it were useless for subjects to have religions it would not be useless for princes to have them." Starting with the premise of non-logical conduct, we are carried to a conclusion directly opposite: the person in command needs rational combinations particularly, and the person who obeys needs more particularly an unreasoned rule independent of his scant knowledge.

365. The weakness in Bayle's argument is not the one that Montesquieu criticizes. It lies in an altogether different direction. After noting and amply demonstrating that human beings do not act according to logical inferences from principles, from opinions, and that a great many human actions of great importance are non-logical, Bayle should have centred his attention upon such actions. Then he would have seen that they were of many kinds; and he would have had to decide whether they were independent or influenced one another mutually. He would readily have seen that they do exert reciprocal influences, and therefore that the social importance of religion lies not at all in the logical value of its dogmas, its principles, its theology, but rather in the non-logical actions that it promotes. He was actually on the road to that conclusion when he asserted that "a religion has to be judged by the cult which it practises"; and when he stated that the pagan religion stopped at a purely external ritualism, he could hardly have been closer to experimental truth. One step more and he would have had the truth entire. But unfortunately he turns aside. Instead of judging religions, which are non-logical actions, by their social influence, he loses his way in questions as to their moral value, or better, as to their relation to what he is pleased to call "morality"; and in that we have a counter-attack by logic, which is again invading territory from which it had been expelled.

From that point of view one might repeat of Bayle what Sumner Maine says of him in commenting on the writings of Rousseau:¹ "It [Rousseau's] was the first attempt to re-erect the edifice of human belief after the purely iconoclastic efforts commenced by Bayle, and in part by our own Locke, and consummated by Voltaire." But that goes to show how, in view of the indefiniteness of ordinary language, utterly different concepts may be expressed in the same words. Maine is thinking not of science or theory but of practice, as is clearly apparent from what immediately follows: "and [Rousseau's system has], besides, the superiority which every constructive effort will always enjoy over one that is merely destructive." It is not the function of theory to create beliefs, but to explain existing ones and discover their uniformities. Bayle took a great step forward in that direction in exposing the vacuity of certain interpretations and opening the way for the discovery of others more consistent with the facts. From the standpoint of theory, his work, far from being inferior to Rousseau's, is as superior to Rousseau's as the astronomy of Kepler is superior to the astronomy of Cosmas Indicopleustes. He may be blamed only for stopping too soon on a road which he had so splendidly opened.

366. Why he did so is hard to guess. The case is not rare. It would seem as though in science it is often necessary to destroy before building can begin. It may also be that Bayle was deterred from a complete expression of his ideas by the moral and religious persecutions common in his time, that the atmosphere of persecution affected the thinker not only materially but intellectually also, and constrained him to disguise his thought under certain forms. Just so in our own time persecutions and annoyances of all sorts emanating from votaries of the religion of sexual virtue have created an atmosphere of hypocrisy in speech and thought that influences writing. And so, if in some future age the expression of human thought comes to be liberated from sex "ties" just as it has already been freed of the ties requiring deference to the Bible, people desiring to understand the thought of writers of our day will have to take account of the masks with which it is disguised in deference to contemporary prejudices. Another cause may have been the scientific inadequacies of ordinary language. If Bayle had not had at his disposal such terms as "religion" and "morality," which seem to be exact but are not, he would have been compelled to deal with things instead of with words, sentiments, fictions; and in that case perhaps he might not have lost his way (§ 114).

367. But his case is merely typical of a vastly populous class of cases where error in argument is directly proportionate to defects in language. Anyone, therefore, desirous of remaining in the logico-experimental field and concerned not to be led astray into the domain of sentiment, must ever be on his watch against this the greatest enemy of science (§ 119). In social matters, human beings as a rule use language that lures them away from the logico-experimental domain. What does such language really mean? We have to be clear on that question before we can go farther, and to it we shall devote the chapter next following.

1. 256 ¹ On the shores of the Lake of Geneva one may see flocks of swans each of which occupies a certain area of the lake. If a swan of one flock tries to invade the territory of another flock, it is attacked, beaten, driven off. The old swans die, young ones are hatched and grow up, and the flock endures as a unit.

2. 257 ¹ Histoire des Romains, Vol. III, p. 411 (Mahaffy, Vol. III, pp. 398-99).

3. 258 ¹ [I read these sentences as interrogations. They are declarative in the original. Evidently the paragraph has been transferred to this point from some place in Chap. I, Pareto neglecting to establish connections.—A. L.]

4. 263 ¹ Pareto, Manuel, pp. 145-46.

5. 263 ² Two very important books on mathematical economics are Osorio's Théorie mathématique de l'échange, and Moret's L'emploi des mathématiques en économie politique.

6. 271 ¹ Politica, I, 1, 4 (Rackham, p. 5): καὶ τοῦτο οὑκ ἑκ προαιρέσεως . . .

7. 271 ² I, 1, 8 (Rackham, p. 9): Διὸ πᾶσα πόλις φύσει ἑστίν, εῐπερ καὶ αἱ πρῶται κοινωνίαι.

8. 272 ¹ Similar contradictions are observable in metaphysical and theological disputes as to "free will," "predestination," "efficacious grace" (§ 280), and the like. Pascal well ridicules some of these incoherences; but, speaking as a metaphysicist and theologian himself, he replaces them with arguments that are worth but little more, and sometimes less. He had begun by saying, Lettres à une provinciale, I, p. 6: "I never quarrel over names, provided I am told what meanings they are given"; and with that he was almost taking his stand within the domain of logico-experimental science (§ 119). But he soon relapses, to go back to the domain of metaphysics, theology, sentiment.

9. 273 ¹ I, 1, 10 (Rackham, p. 11): Οἰθὲν γάρ, ὡς φαμέν, μάτην ἡ φύσις ποιεῖ; Rackham: "does nothing without a purpose."

10. 274 ¹ De lege agraria, II, 35, 95. In combating the Agrarian Law Cicero was trying to persuade his fellow-citizens that a colony established at Capua might become dangerous to Rome. For that reason he may not have been altogether convinced by his own argument. But we need not go into that. We are trying to ascertain not Cicero's personal views, but the opinions current in his time. And if he used the argument he used, it means that he thought it reflected the feeling of a larger or smaller element among Roman citizens.

11. 275 ¹ One may also detect a certain conception of non-logical conduct in the fact that Aristotle ascribes the virtues—temperance, justice, courage, and so on—to the non-rational part of the human being, Magna moralia, I, 5, 1 (Stock, p. 1185-b): "Foresight, intelligence (quickness of wit), wisdom, learning (aptitude for learning), memory, and other similar things arise in the rational part [of the soul]. In the non-rational one finds what are called the virtues: temperance, justice, energy, and all other moral qualities that are deemed worthy of praise." Aristotle's doctrine of the logical or non-logical character of conduct in general was perhaps not very clear—such doctrines rarely are. All the same he seems to have recognized non-logical elements, supplementing them with logical elements, and subordinating them to the logical. In the Politica, VII, 12, 6 (Rackham, p. 601), he says that three things make a man good and virtuous: φύσις, ἔθος, λόγος: "nature, habit, reason." As for the non-logical element, Aristotle admits that human beings act, in part at least, under the influence of external circumstances, such as climate, soil, and so on. In Ibid., VII, 6 (Rackham, pp. 565-66), he clearly relates the conduct of human beings to such circumstances; and in De partibus animalium, II, 4 [An erroneous reference: read: Historia animalium, VIII, 28-29 (Thompson, pp. 606-07).—A. L.], he explains just how he thinks the relationship functions, in general, for living beings. The author (Aristotle ?) of the Problemata, offers, XIV (Forster, pp. 909-10), additional reflections on such relationships. So far we are within the domain of the non-logical. But the writer at once takes steps to be rid of it by a procedure that is general and which lies in subordinating it to logic: it becomes the material with which reason works. Magna moralia, I, 11, 3 (Stock, p. 1187-b): "Judgment, will, and all that is in accord with reason, constitute the principle of conduct, good or bad." Aristotle is not aware that in that he is contradicting what he said, in the Politica, that people who live in cold countries are courageous. In this case, the "principle" of courageous action, that is to say, the "judgment and will" to expose oneself to peril, is determined, according to Aristotle, by climate and not by "reason." He thinks he clears his traces by saying, Magna moralia, I, 11, 5 (Stock, loc. cit.), that first requisite is help from nature, and next will; but ignoring any metaphysical question as to the freedom of the will, which we choose not to go into, we still have the problem, first of knowing whether the two things that he considers independent are so in reality, and then in what proportions they figure in any concrete act. Going into that problem, one finds that there is conduct in which the first element, the non-logical, prevails, and other conduct in which the second element, the logical, prevails.

    Aristotle was lured from the scientific path, aside from metaphysical considerations, by that great enemy of all social science: the mania for achieving some practical result. In the Ethica Nicomachea, II, 2, 1 (Rackham, p. 75), he says that he does not desire to confine himself to theory only: "For we study not to know what virtue is, but to become good; otherwise our study would be of no use." Aristotle had no other means of influencing others than logical argument; and so he was, as he had to be, inclined to make logic the controlling force in human conduct.

12. 278 ¹ De legibus, IV. Aristotle, Politica, VII, 5, also discusses the advantages and disadvantages of proximity to the sea.

13. 278 ² Respublica, III, 21, 415A. And cf. my Systèmes socialistes, Vol. I, p. 276.

14. 280 ¹ Ritter, Geschichte der Philosophie alter Zeit, Vol. III, p. 305 (Morrison, Vol. III, pp. 262-63): "More interested in didacticism than in physic, Socrates sought the principle of all morality strictly in dialectic. So virtue, in his opinion, had no other foundaton than reason and knowledge. But Plato already had found that courage and moderation, two necessary phases of virtue, must pre-exist in the temperament of the human being, whose impulses lie in the heart, not in the head. Aristotle went even farther in that direcdon and clung more tightly still to physic, for which he had a temperamental predilection. As the first principle of virtue he takes not reason but natural impulse and the emotional states of the soul (πάθη)." Zeller, Philosophie der Griechen, Vol. III, p. 118 (missing in Alleyne): [For Socrates] "knowledge is not just an indispensable prerequisite, not just an auxiliary, to true morality: it directly constitutes all morality; and when knowledge is lacking, he is not content with the mere recognition of an imperfect virtue: he cannot see any virtue at all. Not till later on, in Plato, and more completely in Aristotle, will we find a correction of that narrow form of the Socratic doctrine of virtue."

15. 282 ¹ Summa theologiae, I^a II^ae, qu. 55, art. 4 (Opera, Vol. VI, p. 353): "Virtus est bona qualitas seu habitus mentis qua recte vivitur et qua nullus male utitur et quam Deus in nobis sine nobis operatur." The non-logical character of certain conduct is more clearly perceived in a following remark by the Angelic Doctor: "But it should be noted that of the active dispositions (habituum operativorum) some are always towards the bad, such as vicious inclinations; some are now towards the good, now towards the bad, much as opinion stands towards the true and the false."

16. 285 ¹ The Classification of the Sciences, Addendum, pp. 37-38.

17. 286 ¹ Comte is to an extent aware of the evolution he has undergone, Système, Vol. III, Preface, p. vii: "Comparing this volume with the historical portions of my fundamental treatise, it will be noted that my general system is deeper and more complete, whereas my special demonstrations are less developed. From the latter point of view, this final elaboration of my philosophy of history is at variance with my original announcements, which promised more details and proofs in this volume than in my first outlines, to which, instead, I am now obliged to refer for such things. Brought to a clearer understanding of the true character of the philosophical régime, I have come to feel that systematic assertions, which I first regarded as something merely provisory, should be the normal rule of any truly systematic exposition. The progress I have made and the prestige it has won for me allow me in my advancing years to fall in with the free and rapid stride of my chief predecessors, Aristotle, Descartes, and Leibnitz, who simply formulated their thoughts, leaving the task of verifying and developing them to their readers. That division of labour in intercourse between minds is at once the most honourable for the initiated and the most profitable for founders." And in this last, Comte is unquestionably right! It is no little convenience if one can manage to be believed without being pestered for proofs!

18. 287 ¹ Système, Vol. IV, pp. 377-78: "To modify public life, it is enough for [the Priesthood of Humanity] that circumstances shall have brought to the fore some preponderant and responsible will. That condition has been fairly well provided for in France since the advent of the Dictatorship, which frees organized doctrine from the irksome obligation of deferring to legislatures that are ever disposed to perpetuate a revolutionary condition, even when they are reactionary. . . . Without having to convert either the public or its leaders, Positivism, therefore, in virtue of its fundamental truth and its utter seasonableness, can win a partial ascendancy adequate for realizing the final transition, even unbeknown to the principal supporters of the movement." An action that takes place unbeknown to the individual who performs it obviously belongs to the genus of non-logical actions.

19. 289 ¹ Principles of Sociology, Vol. I, § 52.

20. 295 ¹ The Positive Philosophy of Auguste Comte, p. 96 (London, p. 103).

21. 296 ¹ Thebaid, III, v. 661. The scholiast Lactantius [read Luctatius Placidus; see Knaack, Rhenisches Museum für Philologie, Vol. 56, p. 166.—A. L.] annotates [not very keenly] (Leyden, p. 406): "He says that the gods are worshipped for no other reason than the fear of mortals. As Lucan says, Pharsalia, I, v. 486: 'They fear inventions of their own devising' (quae finxere timent). Petronius [Fragmenta, XXVII] follows Statius: 'Fear first created gods on earth.' And Mintanor Musicus writes: '. . . the gods, whom humanity first invented under sting of pain.' "

22. 296 ² Holbach, Système de la nature, Vol. I, pp. 448, 456: "Mankind has ever derived its basic ideas on divinity from ignorance, fear, and calamity. . . . Man's earliest theology taught him first to fear and worship the elements themselves, and crude material objects."

23. 296 ³ We noted Cicero's view of the practices of Roman divination in § 182^a: De divinatione, I, 3, 3: "Atque haec, ut ego arbitror, veteres rerum magis eventis moniti quam ratione docti probaverunt." That is very often the case: the fact, the non-logical action, comes first, then the explanation of the fact, the logical varnish.

24. 298 ¹ The reservations relate to Mill's not very exact use of terms such as "legitimate" and "illegitimate." But Mill cannot be specially blamed for that. It is a defect common to almost all writers who deal with such subjects.

25. 299 ¹ Mill, innocent soul, goes on to say, loc. cit.: "To justify that [such constraint], the conduct from which it is desired to deter him must be calculated to produce evil to someone else." He did not realize that sophistries are never wanting to show that the damage is there. Notice what happens in countries where people set out to enforce temperance and virtue in the holy name of "Progress": Giornale d'ltalia, March 19, 1912: "Atlanta, Georgia, March 2. Last evening Commendatore Alessandro Bonci, who was stopping here temporarily in connexion with professional engagements, was arrested at the Georgian Terrace Hotel, together with his wife, his secretary, and his pianist, for violating the liquor law. It seems that Signor Bonci and his friends, like good Italians, who serve wine two meals a day at least, had adopted an ingenious device for doing so in spite of the law that forbids the use of wines and liquors in the State of Georgia. For several days the manager of the hotel had noticed that towards the middle of their meals the Boncis and their friends were in the habit of setting on the table four little bottles such as are used by druggists, with labels giving directions for using the presumptive 'medicines.' The regularity with which the Bonci party drank the contents of the bottles twice a day, as though each member of it were suffering from the same disease and required the same treatment, at length aroused the suspicions of the house detective. He mentioned the matter to a zealous policeman, who last evening, when the time for the 'treatment' came, confiscated the bottles. Each of them was found to have the capacity of a wine-glass and to contain nothing but excellent Chianti, with which, it seems, Commendatore Bonci travels well supplied in order to cope with the surprises of American law. Despite the lively protestations of Signor Bonci, the four offenders were put into an automobile and taken to the Court House, where Judge Ralendorf, after a summary inquiry, continued the case till this morning, fixing bail at $2,000. Then came the best, not to say the worst, of it. The celebrated tenor found he had no more than $150 in his pocket, and he was faced with the prospect of spending the night in jail."

    We may guess that if Signor Bonci had remembered that the ointment of St. John Goldmouth may be used on the hands of American reformers with as good effect as it had in Boccaccio's time on the hands of our virtuous Italian Inquisitors, he might have escaped such annoyance. In general terms: You happen to be in the dining-car when the train enters one of the abstemious states of the American Union, and the glass of wine that you were about to drink is snatched from the table in front of you. If you ask, "What harm am I doing to my neighbour by drinking this glass of wine?", the answer comes quick and prompt: "You are setting a bad example!" And the rabble that enforces its will upon you in that fashion speaks with indignation of Spanish Catholics who, to prevent setting bad examples, refuse to tolerate in Spain any public worship except the Roman Catholic!

26. 301 ¹ Esquisse d'un tableau historique des progrès de l'esprit humain, pp. 264-65.

27. 303 ¹ Système de la nature, Vol. I, pp. 398-409: "The errors of mankind as to what constitutes happiness are the real source of its troubles. Inefficacy of proposed remedies. . . . If we consult experience, we see that the real source of that multitude of woes that everywhere afflict the human race is to be sought in sacred opinions and illusions. Ignorance of natural causes first created gods for humanity: imposture clothed them with terror. The deadly thought of them pursued the human being without making him better, filled him with fears to no purpose, packed his mind with nightmares, blocked the progress of his intelligence, prevented him from seeking his own welfare. His fears enslaved him to deceivers who made pretence of working his weal. . . . Prejudices no less dangerous have blinded men as to their rulers. . . . A similar blindness we find in the science of morals. . . . So humanity's burden of woe has no whit been lightened, but has been made heavier rather by his religions, his governments, his education, his opinions, in a word by all the institutions that he has been persuaded to adopt [By whom persuaded? By someone not of the human species?] on pretence that his lot would be made more bearable. It cannot be too often repeated: In error lies the true source of the ills that afflict the human race. Not with Nature lies the responsibility for human unhappiness. No angry God ever willed that humanity should live in tears. No hereditary depravity made mortals wicked and miserable. Those deplorable consequences are all and exclusively due to error."

28. 303 ² Elie Reclus, Les primitifs, p. 161: "Since morality is measured, along its general lines at least, by intellectual development, no surprise will be occasioned by finding it very rudimentary here [among the Redskins]."

29. 305 ¹ Farther along, in Chapter IX, we shall have to consider a still more general subject—the variability of the arguments to which human beings are prompted by sentiments, and which provide logical exteriors for non-logical conduct. A strictly inductive course, such as we have been following, brings up the particular problem in advance of the general. That has the drawback of compelling us to examine the particular problem first, and to keep going back to things on which we have already touched. It has, on the other hand, the great advantage of making the materials we work with clearer and more manageable.

30. 306 ¹ [One need hardly remind the reader that these synoptic pictures of Pareto's classifications are unintelligible apart from the exposition seriatim of the various categories that he proceeds to make. They have to be continually re-read in connexion with the text that follows. This table is particularly obscure in itself, not only because of exceptionally opaque writing but because implicit in it is another classification that Pareto for some reason does not see fit to utilize. It is clear that the devices in Class A are used from a sceptical standpoint to discredit beliefs on logical grounds. The B-I and B-III devices are used by believers to represent their beliefs as logical. The other devices are "errors" commonly made by scholars in viewing the non-logical as logical. I use the term "device" for the sake of clarity; Pareto's term was "means." Whatever the term used, it has to be understood as not implying any intent to deceive on the part of a person using such a device or means.

    Pareto's classifications, which are taken over from botany, envisage classes, genera and subgenera (sometimes species and subspecies). I keep these terms in the tables of classification. In the text at large, to avoid a fatiguing technical atmosphere, I often render "genus" and "species" loosely as "type," "kind," "sort," or more generally "variety": the "Iβ variety," or "1β type" would be, in the tables, the "1β subgenus," and so on. Pareto makes but little use of the "genus" in the structure proper of his theories, the one exception perhaps being his analysis of the residue of asceticism (§§ 1163 f.). The "class," on the other hand, is essential to his theory of interdependence and intensities (Chapter XII). Since residues increase or diminish in intensities by "classes," and interdependences arise primarily within "classes," it is clear that the structure of the "class" has all along to be borne in mind.—A. L.]

31. 306 ² "Principle" here means the cause to which an action is to be ascribed.

32. 307 ¹ Phaedrus, 229-30 (Fowler, pp. 419-23).

33. 307 ² The fact is mentioned by Xenophon's Socrates. Memorabilia, IV, 3, 16: "Since thou seest that when the god of Delphi is asked how best to please the gods, he replies: By following the custom of the city." Cicero, De legibus, II, 16, 40: "Our law shall further provide that of all our ancestral rites the best should be fostered. When the Athenians consulted the Pythian Apollo as to which rites they had better practise, they received the oracle: 'Those customary with the forefathers.' Then they came back again, saying that the custom of the forefathers had often changed, and they asked which they should prefer of the various ancestral customs; and the god answered: 'The best.' " Cicero appends a logical consideration that has no logical force whatever: "And it is assuredly true that what is best should be taken as the most authentic tradition and the closest to God."

34. 308 ¹ III, 2, 5. Cf. De divinatione, II, 12, 28: "As regards divination, I think the custom should be cherished for considerations of state and common religion. But here we are in strict privacy and we surely have a right to discuss the matter quite frankly (sine invidia), and I in particular, since I have very grave doubts in not a few connexions."

35. 308 ² [Pareto wrote: "which means either that such [logico-experimental] reality does not exist or that it is of the genus of the principles of non-logical actions."—A. L.]

36. 310 ¹ Remarques pour servir de Supplément à l'Essai sur les mœurs, Pt. IV (Œuvres, Vol. V, p. 48): "Contemptible customs in a nation do not always indicate that that nation is itself contemptible." Among the blunders mentioned are the following: 1. Cicero's essay De divinatione was written after Caesar's death. But that is a small matter; if one is going to pretend that Caesar sent ambassadors to China, one may also pretend that he was living when Cicero wrote the essay. 2. The Chinese pantheon was much better filled than the Roman pantheon. That error on Voltaire's part may be forgiven, since it was the error of all the philosophes of his time. With a little care, however, he might have avoided the following: 3. Wittingly or unwittingly, he confuses Roman divination with the Etruscan. The god Tages belonged only to the latter. 4. Jupiter Optimus Maximus was by no means the only god in the official cult of Rome. [I cannot believe that Voltaire did not know that. The very glaringness of the error calls attention to a sacrilegious parody of French Chrisdanity in the allusions to Jupiter, Flora, and the Penates.—A. L.] 5. The Penates were not at all the gods of silly housewives. Servius, In Vergilii Aeneidem, II, v. 514 (Thilo-Hagen, Vol. I, p. 298): "The Penates are all the gods worshipped in the home." Rome herself had her Penates. Voltaire would use Cicero against the silly housewives, but Cicero himself invokes the Penates, Pro Publio Sulla, 31, 86: "Wherefore, O ye gods of our forefathers, and ye, O Penates, who watch over this city and this country of ours, ye who during my consulship did confer your aid and your divine protection upon this state, upon the Roman People and its liberties, upon these homes, these temples, you do I invoke as witnesses to my integrity and honesty of purpose in appearing in defence of Publius Sulla." Cf. also In Catilinam, IV, 9, 18. 6. Whether he believed in such things or not, Caesar made a practice of consulting soothsayers. There is an allusion to that in De divinatione, I, 52, 119; II, 16, 36, which Voltaire quotes; and cf. Dio Cassius, Historia Romana, XLIV, 17, 18; Plutarch, Caesar, 63-64 (Perrin, Vol. VII, pp. 589-95); Suetonius, Divus Julius, 81; Pliny, Historia naturalis, XXVIII, 4 (2). To one of Cæsar's superstitions we have previously alluded in § 184². 7. Cicero does not dream of ridiculing all auspices. He was himself an augur, and speaks of auspices with the greatest respect, De legibus, II, 12, 31: "The office of augur stands very high and is of the greatest importance in the state [i.e., in Cicero's ideal state] and it is clothed with the greatest prestige. And that I feel not because I am an augur but because we can think not otherwise." He had little or no regard for the intrinsic merits of augury; but he considered the institution useful to the state and consequently did not ridicule it (cf. the quotations in § 313¹). 8. Cato was speaking not of the augurs, but of the haruspices: Cicero, De divinatione, II, 24, 51: "Familiar the old jest of Cato, who used to express his wonder that one haruspex could ever look at another without laughing." For that matter it is a common error to confuse Roman augury with Etruscan divination by inspection of entrails. Only when they could not help doing so did the Romans appeal to Etruscan divination. Tiberius Gracchus, the father of the Gracchi, on being accused by Etruscan soothsayers, who were functioning at an election, of calling for a vote against the auspices, addressed them as follows: Cicero, De natura deorum, II, 4, 11: " 'You say that I am not in order, though I am putting this question as consul and as augur, and under good auspices? And you, Etruscans, you, barbarians—you presume to say what good auspices for the Roman People are? You presume to be interpreters for these comitia?' And he bade them to be gone from the Forum."

37. 311 ¹ Strictly speaking, this remark and the next following are irrelevant to the present chapter. I make them simply to warn anew of the habit people have of assuming that a writer says what he does not say (§§ 41, 74-75).

38. 312 ¹ If one were to say "kept," or "preserved," instead of "invented" in the proposition in question, it would at times correspond to a greater or lesser extent with reality (§ 316).

39. 313 ¹ Historiae, VI, 56, 8-12 (Paton, Vol. III, p. 395). After noting the great rôle of religion in Roman public life, Polybius adds: "That will seem strange to many. As for me, I believe that religion was established with an eye to the masses. In fact, if the city were made up entirely of educated people, such an institution might never have been called for. But since the masses everywhere are fickle and untrustworthy, full of lawless passions, unreasoning angers, violent impulses, they can be controlled only by mysterious terrors and tragic fears. It seems to me, therefore, that not by chance and not without strong motive did the ancients introduce these beliefs in gods and hells to the multitude." Strabo, Geographica, I, 2, 8 (Jones, Vol. I, p. 71): "Since neither women in the mass nor the utterly untutored mob can be influenced by philosophical discourse and preached into piety, reverence, and faith, superstition has to be called in." And then: ". . . myths being like that and turning out to the advantage of society, civilized living, and the continuity of the human race." Cf. Plutarch, Adversus Colotem, 31 (Goodwin, Vol. V, pp. 379-80). Then Livy, Ab urbe condita, I, 19, 4: "He [Numa] thought that fear of the gods should be instilled the very first thing, as a most effective measure for a populace that in those days was still crude and ingenuous (imperitam)." Here we are wholly within the realm of logical conduct, the masses being lured into religion by subterfuge. Cicero, De legibus, II, 13, 32 (Atticus, alluding to the different views of the two augurs Marcellus and Appius): " 'I have examined their writings and I find that according to the one, the auspices you mention were devised for purposes of state; while according to the other it would seem that you can actually foretell the future by your science.' " Cicero, De divinatione, II, 18, 43: "We find it written in our augural commentaries: 'It is sacrilege to hold comitia with Jove thundering or lightning.' That may have been devised for purposes of state, for our forefathers wanted to have some pretext for not holding comitia." Ibid., II, 33, 70: "Yet I believe that Romulus, who founded the city in obedience to auspices, must have thought that there was a science of augury for foretelling the future (antiquity erred in many matters) and we see that that belief has remained unshaken whether by experience, by learning, or by time. However, the custom and science of divination, the strict observance of it, and the prerogatives of the augurs and the prestige of their college, have been kept alive in deference to popular feelings, and in view of their great advantage to the state." A little later, II, 35, 75, he adds that he believed "the augural law to have been first established through belief in divination and to have been kept and preserved later on for reasons of state." That seems to have been Cicero's own opinion and it does not come far from the truth. Non-logical actions arise spontaneously. They may then be kept in deference to tradition or because of their proved usefulness. Of course any logical origin, by design of Romulus, is pure myth. Cf. Aristotle, Metaphysica, XI, 8, 13 (Ross, p. 1074b). After discussing the divinity of the stars, he adds: "The rest is a mythical adjunct, designed to influence the multitude and promote obedience to law and the common welfare." See further: Plutarch, De placitis philosophorum, I, 7, 2 (Goodwin, Vol. III, p. 119); and Sextus Empiricus, Contradictiones, IX, Adversus physicos, II, De diis, 14-16 (551) (Opera, Vol. II, pp. 539-40).

40. 313 ² VI, 11. He is comparing the republic of Lycurgus with the Roman Republic. He believes that Lycurgus was a real person and founded his state with preconceived purposes. Then he goes on: "The Romans achieved the same end in creating their own republic. Not through speculation (οὐ μὴν διὰ λόγου), but through their schooling in many struggles and vicissitudes and through their unfailing choice of what was best did they achieve the same end as Lycurgus and create the best of our governments."

41. 314 ¹ Dissertation sur la politique des Romains dans la religion, p. 303.

42. 317 ¹ Evangelica praeparatio, V (Opera, Vol. III, pp. 307-402).

43. 318 ¹ Dialogo dei due massimi sistemi del mondo, Giornata prima (Opera, Vol. VII, p. 35).

44. 319 ¹ This extreme case recognizes non-logical actions for what they are and therefore ought not, strictly speaking, to be classified with procedures for giving non-logical actions the semblance of logic. However, we must consider it as the point of departure for many such procedures, and so glance at it here.

45. 321 ¹ The sanctionless precept is not of this variety because it does not evade but recognizes the fact that an action is non-logical—indeed it is in the sanctionless precept that non-logical actions can be most readily identified.

46. 322 ¹ Cultes, mythes et religions, Vol. I, pp. 1-2.

47. 322 ² We have here been considering the sanction appended to taboos as a device for logicalizing non-logical actions. Farther along we shall examine them as devices for inducing observance of taboos.

48. 323 ¹ History of Sumatra, p. 250.

49. 323 ² For other examples of the kind see §§ 1430 f.

50. 330 ¹ For fuller explanation see Chapter IX (§§ 1480 f.).

51. 333 ¹ Odyssey, VI, vv. 207-08: πρὸς γὰρ Δισς εἰσιν ἅπαντες ξεῖοί τε πτωχοί τε.

52. 335 ¹ For the detailed argument see §§ 1510 f.

53. 336 ¹ As for the God of the Hebrews, see Piepenbring, Théologie de l'Ancien Testament, pp. 98-99: "The holiness of God is intimately bound up with His jealousy, His wrath, His vengeance. . . . In the 'Old Canticle' (Ex. 15:7) Moses cries out to the Lord: '. . . In the greatness of thine excellency thou hast overthrown them that rose up against thee: thou sentest forth thy wrath, which consumed them as stubble.' [Can any neo-Chrisdan abstraction say as much?] The wrath of God breaks out in the form of dire punishment every time His will is crossed, disregarded, transgressed." These milk-and-water Christians are inclined to think that all that changed with the coming of Christ, but such is not the case. The early Church Fathers discourse without mincing words on the punishments that will be visited on unbelievers. As for the God of St. Paul, one of the many passages will suffice: I Cor. 10:8: "Neither let us commit fornication, as some of them [the Israelites] committed, and fell in one day three and twenty thousand [Num. 25:1-9]." Can the abstraction concocted by the pseudo-science of the neo-Christians pretend to do as much? No! Well, in that case the precept will be obeyed by those who are already good Christians, and no one who is not will pay any attendon. But that is the essential characteristic of the principles (§ 306¹) of non-logical actions. The Apostle continues: "Neither let us tempt Christ, as some of them also tempted, and were destroyed of serpents [Num. 21:4-9]. Neither murmur ye, as some of them also murmured, and were destroyed of the destroyer [Num. 11:16]." And later on, 22, he asks: "Do we provoke the Lord to jealousy? are we stronger than He?" Every sensible man will answer no if the being in question is an omnipotent God; but many sensible men will answer yes if it is a question of an abstraction that some few individuals have distilled from their own sentiments.

54. 337 ¹ Auguste Sabatier, Les religions d'autorité et la religion de l'esprit, pp. 440-41 (English translation, pp. 281-82): "The letter, the alphabetic sign, characterizes the Mosaic religion in accordance with the form of its appearance in history, its manner of being and action. . . . The letter kills. Spirit, instead, characterizes the religion of the Gospel in accord with the very nature of the inner moral relationship that it sets up between God and man, in accord with the manner of being of the Gospel and the principle of its action. . . . In view of that you must surely understand what the religion of the spirit is. It is the religious relationship realized in pure spirituality. It is God and man conceived both as spirit and as reciprocally permeating each other to the point of attaining complete communion. Physical bodies are by definition impenetrable to each other. . . . Quite otherwise the relationship between spirits. Their inward tendency is to live each other's lives mutually and to combine in a higher common life. What the law of gravitation is to the physical world as regards the maintenance of its harmony, so love is and so love does in the spiritual and moral world. [The conception this gentleman has of the law of gravitation would make a story.] . . . Ultimate force in the moral development of the human being, the spirit of God no longer constrains him from without but determines, animates, him from within, and is the source of his life. . . . The fulfilment of natural duties, the regular exercise of all human faculties, the progress of enlightenment as well as of justice—that is the perfection of the Christian life. Becoming an inner reality, a fact of conscience, Christianity is now nothing more than conscience raised to its highest power."

55. 346 ¹ Brachet, Grammaire historique de la langue française, pp. 293-94 (Kitchin, pp. 195-96): "Before attaining the degree of exactness that it possesses today, etymology, like all the sciences and perhaps more notably than any other, traversed a long period of infancy, of gropings, of uncertain efforts, during which arbitrary associations, superficial analogies, reckless combinations, made up virtually its whole patrimony." Here Brachet quotes from Réville, Les ancêtres des européens: " 'Abidingly famous the day-dreams of Plato in the Cratylus, the absurd etymologies of Varro [Etymologiae, Dordrecht, Part III, pp. 165-176] and Quintilian among the Romans, the philological fancies of Ménage in France in the seventeenth century. People saw nothing strange about connecting jeûne, "fast," with jeune, "young." Is not youth the morning of life, and is one not fasting when one gets up? Most often two words of entirely different forms were derived from each other, the gulf between them being bridged by fictional intermediaries. That was the way Ménage got the French rat from the Latin mus, "mouse": "People must have said first mus, then muratus, then ratus, finally rat." It was courageously assumed that an object could get its name from a quality opposite to its own, affirmation provoking negation, so that Latin lucus, "grove," came from non lucere, "not to be bright," because on entering a grove one finds it shady.' " Brachet continues: "From such a mass of erudite nonsense how could one of the leading sciences eventually arise in our day? By the discovery and application of the comparative method, which is the method of the natural sciences"—and the method we are trying to follow in these volumes.

56. 349 ¹ For examples see §§ 733 f.

57. 350 ¹ For several such interpretations see Chapter V.

58. 351 ¹ Chips from a German Workshop, Vol. II, p. 64. [The French translation which Pareto used for this passage has a number of errors.—A. L.]

59. 354 ¹ History of Civilization in England, Vol. I, pp. 179-82.

60. 354 ² Buckle quotes James Mackintosh, Condorcet, and Kant, in support.

61. 355 ¹ Pareto, Manuale, Chap. I, § 18.

62. 355 ² [Fielding, Tom Jones, IX, iii, 2: ". . . purposes . . . which though tolerated in some Christian countries, connived at in others and practised in all . . . are expressly forbidden . . . by that religion which is universally believed in in those countries."—A. L.]

63. 356 ¹ Here and there in his work Buckle himself ends by making at least implicit allusion to non-logical actions. Trying to account for the differences between the Puritan Revolution and the French Fronde, he suggests, Vol. II, p. 150, "that in England a war for liberty was accompanied by a war of classes, while in France there was no war of classes at all"; and further, Vol. II, p. 162, that "the object of the [French] nobles was merely to find new sources of excitement, and minister to that personal vanity for which, as a body, they have always been notorious." Now whatever the route that is tried in order to get from such facts to logical inferences from an "intellectual principle," it is certain that the facts depend on natural inclinations, which cannot be regarded as resulting from any differences between the scientific and intellectual attainments of the English and the French at that period. No such differences existed.

64. 358 ¹ Pensées diverses, § 138.

65. 360 ¹ Continuation des Pensées diverses, § 139.

66. 363 ¹ L'esprit des lois, XXIV, 2: Paradoxe de Bayle. Montesquieu was right in saying that "in order to attenuate the horrors of atheism" Bayle was "too severe on idolatry"; but he should have recognized Bayle's artifice in doing that. It was a trick he used himself on other occasions.

67. 365 ¹ Ancient Law, p. 84.