Chapter1: Introduction: Sociological Theory

SUMMARY

Max Weber has had a more powerful positive impact on a wide range of sociological theories than any other sociological theorist. This influence is traceable to the sophistication, complexity, and sometimes even confusion of Weberian theory. Despite its problems, Weber's work represents a remarkable fusion of historical research and sociological theorizing.

We open this chapter with a discussion of the theoretical roots and methodological orientations of Weberian theory. We see that Weber, over the course of ins career, moved progressively toward a fusion of history and sociology, that is, toward the development of a historical sociology. One of his most critical methodological concepts is verstehen. Although this is often interpreted as a tool to be used to analyze individual consciousness, in Weber's hands it was more often a scientific tool to analyze structural and institutional constraints on actors. We also discuss other aspects of Weber's methodology, including his propensity to think in terms of causality and to employ ideal types. In addition, we examine his analysis of the relationship between values and sociology.

The heart of Weberian sociology lies in substantive sociology, not in methodological statements. Although Weber based his theories on his thoughts about social action and social relationships, his main interest was the large-scale structures and institutions of society. We deal especially with his analysis of the three structures of authority legal, traditional, and charismatic. In the context of legal authority, we deal with his famous ideal-typical bureaucracy and show how he used that tool to analyze traditional and charismatic authority. Of particular interest is Weber's work on charisma. Not only did he have a clear sense of it as a structure of authority, he was also interested in the processes by winch such a structure is produced.

Although his work on social structures--such as authority is important, it is at the cultural level, in his work on the rationalization of the world, that Weber's most important insights lie. Weber articulated the idea that the world is becoming increasingly dominated by norms and values of rationalization. In this context, we discuss Weber's work on the economy, religion, law. the polity, the city, and art forms. Weber argued that rationalization was sweeping across all these institutions in the West, whereas there were major barriers to this process in the rest of the world.

Weber's thoughts on rationalization and various other issues are illustrated in his work on the relationship between religion and capitalism. At one level, this is a series of studies of the relationship between ideas (religious ideas) and the development of the spirit of capitalism and, ultimately, capitalism itself. At another level, it is a study of how the West developed a distinctively rational religious system (Calvinism) that played a key role in the rise of a rational economic system {capitalism}. Weber also studied other societies, in winch he found religious systems (for example. Confucianism, Taoism, and Hinduism) that inhibit the growth of a rational economic system. It is this kind of majestic sweep over the history of many sectors of the world that helps give Weberian theory its enduring significance.

Chapter8: Vilfredo Pareto: the Sound from Italy

On Logical and Non-Loglcal Action

Residue: Combinations and Group Persistence

SINCE SOCIAL PHENOMENA
Vilfredo Pareto: A Biographical Sketch

Vilfredo Pareto(1848-1923) An Italian economist and sociologist, known for his application of mathematics to economic analysis and for his theory of the division of logical action and non-logical action. Born in 1848, the son of a Genoese father and a French mother, Pareto studied engineering at the University of Turin. The five-year course in civil engineering, the first two years of which were devoted to mathematics, deeply influenced Pareto's future intellectual outlook. In 1870 he graduated with a thesis on "The Fundamental Principles of Equilibrium in Solid Bodies." His later interest in equilibrium analysis in economics and sociology is prefigured in this thesis. From 1870 until 1893 he worked as an engineer (like his father). Residing in Florence, he studied philosophy and politics and wrote many periodical articles in which he was one of the first to analyze economic problems with mathematical tools. In 1893 he was chosen to succeed ~on Walras in the chair of political economy at the University of Lausanne, Switzerland. He died in 1923 in Geneva.

Pareto's first work, Cours d'economie politique (1896-97), included his famous "law" of income distribution, a complicated mathematical formulation in which he attempted to prove that the distribution of incomes and wealth in society is not random and that a consistent pattern appears throughout history, in all parts of the world and in all societies.

In his Manuale di economia politica (1906) and Manuale d'economie politique (1909—a translation of the preceding item but with a completely redone mathematical appendix) he further developed his theory of pure economics. In this book he laid the foundation of modern welfare economics with his concept of the so-called Pareto optimum, stating that the optimum allocation of the resources of a society is not attained so long as it is possible to make at least one individual better off in his own estimation while keeping others as well off as before in their own estimation.

His most important sociological writing is The Mind and Society (1916). In the book Pareto divided two actions of human being: the logical and non-logical. For Pareto, if economics was the science of logical actions, sociology was the science of non-logical actions. According Raymond Boudon, Pareto's definition of sociology is very important to the extent that it expresses in depth one of the fundamental intentions of sociologists to analyse and explain actions and, more generally, behaviour which seem irrational to the observer.

On Logical and Non-Loglcal Action

THIS is the first step we take along the path of induction . If we were to find, for

instance, that all human actions.corresponded to logico-experimental theories,

or that such actions were the most important, others having to be regarded as phenomena of social pathology deviating from a normal type, our course evidently would be entirely different from what it would be if many of the more important human actions proved to correspond to theories that are not logico-experimental.

Let us accordingly examine actions from the standpoint of their logico-experimental character. But in order to do that we must first try to classify them, and in that effort we propose to follow the principles of the classification called natural in botany and zoology, whereby objects on the whole presenting similar characteristics are grouped together. In the case of botany Tournefort's classification was wisely abandoned, It divided plants into "herbs" and "trees," and so came to separate entities that as a matter of fact present close resemblances. The so-called natural method nowadays preferred does away with all divisions of that kind and takes as its norm the characteristics of plants in the mass, putting like with like and keeping the unlike distinct. Can we find similar groupings to classify the actions of human beings?

It is not actions as we find them in the concrete that we are called upon to classify, but the elements constituting them. So the chemist classifies elements and compounds of elements, whereas in nature what he finds is mixtures of compounds. Concrete actions are synthetic- they originate in mixtures, in varying degrees, of the elements we are to classify.

Every social phenomenon may be considered under two aspects: as it is in reality, and as it presents itself to the mind of this or that human being. The first aspect we shall call objective, the second subjective. Such a division is necessary, for we cannot put in one same class the operations performed by a chemist in his laboratory and the operations performed by a person practising magic the conduct of Greek sailors in plying their oars to drive their ship over the water and the sacrifices they offered to Poseidon to make sure of a safe and rapid voyage. In Rome the Laws of the XII Tables punished anyone casting a spell on a harvest. We choose to distinguish such an act from the act of burning a field of grain.

We must not be misled by the names we give to the two classes. In reality both are subjective, for all human knowledge is subjective. They are to be distinguished not so much by any difference in nature as in view of the greater or lesser fund of factual knowledge that we ourselves have. We know, or think we know, that sacrifices to Poseidon have no effect whatsoever upon a voyage. We therefore distinguish them from other acts which (to our best knowledge, at least) are capable of having such effect. If at some future time we were to discover that we have been mistaken, that sacrifices to Poseidon are very influential in securing a favourable voyage, we should have to reclassify them with actions capable of such influence. All that of course is pleonastic. It amounts to saying that when a person makes a classification, he does so according to the knowledge he has. One cannot imagine how things could be otherwise.

There are actions that use means appropriate to ends and which logically link means with ends. There are other actions in which those traits are missing. The two sorts of conduct are very different according as they are considered under their objective or their subjective aspect. From the subjective point of view nearly all human actions belong to the logical class. In the eyes of the Greek mariners sacrifices to Poseidon and rowing with oars were equally logical means of navigation.To avoid verbosities which could only prove annoying, we had better give names to these types of conduct. Suppose we apply the term logical actions to actions that logically conjoin means to ends not only from the standpoint of the subject performing them, but from the standpoint of other persons who have a mere extensive knowledge--in other words, to actions that arc logical both subjectively and objectively in the sense just explained. Other actions we shall call nonlogical (by no means the same as "illogical"). This latter class we shall subdivide into a number of varieties.

A synoptic picture of the classification will prove useful:

GENERA AND SPECIES HAVE THE ACTIONS LOGICAL

ENDS AND PURPOSES:

Objectively? Subjectively?

CLASS Ⅰ. LOGICAL ACTIONS

(The objective end and the subjective purpose are identical. )

Yes Yes

CLASS Ⅱ. NON-LOGICAL ACTIONS

(The objective end differs from the subjective purpose. )

Genus 1 No No

Genus 2 No Yes

Genus 3 Yes No

Genus 4 Yes Yes

SPECIES OF THE GENERA 3 AND 4

3α, 4α The objective end would be accepted by the subject if he knew it.

3β, 4β The objective end would be rejected by the subject if he knew it.

The ends and purposes here in question are immediate ends and purposes. We choose to disregard the indirect. The objective end is a real one, located within the field of observation and experience, and not an imaginary end, located outside that field. An imaginary end may, on the other hand, constitute a subjective purpose.

Logical actions are very numerous among civilized peoples. Actions connected with the arts and sciences belong to that class, at least for artists and scientists. For those who physically perform them in mere execution of orders from superiors, there may be among them non-logical actions of our II-4 type. The actions dealt with in political economy also belong in very great part in the class of logical actions. In the same class must be located, further, a certain number of actions connected with military, political, legal, and similar activities.

So at the very first glance induction leads to the discovery that non-logical actions play an important part in society. Let us therefore proceed with our examination of them.

First of all, in order to get better acquainted with these non-logical actions, suppose we look at a few examples. Many others will find their proper places in chapters to follow. Here are some illustrations of actions of Class II:

Genera 1 and 3, which have no subjective purpose, are of scant importance to the human race. Human beings have a very conspicuous tendency to paint a varnish of logic over their conduct. Nearly all human actions therefore work their way into genera 2 and 4. Many actions performed in deference to courtesy and custom might be put ill genus 1. But very very often people give some reason or other to justify such conduct, and that transfers it to genus 2. Ignoring the indirect motive involved in the fact that a person violating common usages incurs criticism and dislike, we might find a certain number of actions to place in genera t and 3.

Says Hesiod:"Do not make water at the mouth of a river emptying into the sea, nor into a spring. You must avoid that. Do not lighten your bowels there, for it is not good to do so." The precept not to befoul rivers at their mouths belongs to genus 1. No objective or subjective end or purpose is apparent in the avoidance of such pollution. The precept not to befoul drinking-water belongs to genus 3. It has an objective purpose that Hesiod may not have known, but which is familiar to moderns: to prevent contagion from certain diseases.

It is probable that not a few actions of genera 1 and 3 are common among savages and primitive peoples. But travellers are bent on learning at all costs the reasons for the conduct they observe. So in one way or another they finally obtain answers that transfer the conduct to genera 2 and 4.

Another very important difference between human conduct and the conduct of animals lies in the fact that we do not observe human conduct wholly from the outside as we do ill the case of animals. Frequently we know the actions of human beings through the judgments that people pass upon them, through the impressions they make, and in the light of the motives that people are pleased to imagine for them and assign as their causes. For that reason, actions that would otherwise belong to genera 1 and 3 make their way into 2 and 4.

Operations in magic when unattended by other actions belong to genus 2. Thle sacrifices of the Greeks and Romans have to be classed in the same genus--at least after those peoples lost faith in the reality of their gods. Hesiod, Opera et dies, vv. 235 - 39, warns against crossing a river without first washing one's hands in it and uttering a prayer. That would be an action of genus 1. But he adds that the gods punish anyone who crosses a river without so washing his hands, That makes it an action of genus 2.

This rationalizing procedure is habitual and very wide-spread. Hesiod says also, that grain should not be sown on the thirteenth of a month, but that that day is otherwise very auspicious for planting, and he gives many other precepts of the kind. They all belong to genus 2.In Rome a soothsayer who had observed signs in the heavens was authorized to adjourn the comitia to some other day. Towards the end of the Republic, when all faith in augural science had been lost, that was a logical action, a means of attaining a desired end. But when people still believed in augury, it was an action of genus 4. For the soothsayers who, with the help of the gods, were so enabled to forestall some decision that they considered harmful to the Roman People, it belonged to our species 4a, as is apparent if one consider that in general such actions correspond, very roughly to be sure, to the provisions used in our time for avoiding ill-considered decisions by legislative bodies: requirements of two or three consecutive readings, of approvals by two houses, and so on.

Most acts of public policy based on tradition or on presumed missions of peoples or individuals belong to genus 4.William I, King of Prussia, and Napoleon Ill, Emperor of the French, both considered themselves "men of destiny." But William I thought his mission lay in promoting the welfare and greatness of his country, Louis Napoleon believed himself destined to achieve the happiness of mankind. William's policies were of the 4a type; Napoleon's, of the 4B.

Human beings as a rule determine their conduct with reference to certain general rules (morality, custom, law), which give rise in greater or lesser numbers to actions of our 4a and even 4B varieties.

Logical actions are at least in large part results of processes of reasoning. Non-logical actions originate chiefly in definite psychic states, sentiments, subconscious feelings, and the like. it is the province of psychology to investigate such psychic states. Here we start with them as data of fact, without going beyond that.

Residue: Combinations and Group Persistence

SINCE SOCIAL PHENOMENA appear in complex form in the concrete, we saw at once that it would be helpful to divide them into at least two elements, distinguishing logical from non-logical conduct; and that gave us a first conception of the nature of non-logical conduct and of its importance in human society. But at that point a question arose: If non-logical conduct plays such an important role in human life, why has it been so generally neglected? We found in reply that almost all writers on social or political subjects have indeed observed such conduct, or at least caught glimpses of it. Many elements, therefore, of the theory we are framing in these volumes are to be found scattered about here and there in the works of various writers, though often under hardly recognizable forms.

But we saw that all such writers had ideas of their own to which they very expressly attached capital importance-ideas on religion, morality, law, and the like, which have been battle-grounds for centuries. So, if they did recognize non-logical conduct implicitly, explicitly they glorified logical conduct, and most of them regarded it as the only conduct worth considering in social phenomena. We were therefore called upon to see what truth there was in theories of that type, and to decide whether we were to abandon the course on which we had set out or take heart and push on.

We then proceeded to examine those various manners of considering social phenomena, and we saw that from the logico-experimental standpoint they were devoid of all exactness and of any strict accord with the facts~ though from another standpoint, we could not deny the great importance that they had had in history and in determining the social equilibrium. That discovery lent force to a suspicion which had already occurred to us, and which will acquire greater and greater prominence in the course of these volumes: that the experimental "truth" of certain theories is one thing and their social "utility" quite another, and that the two things are not only not one and the same but may, and often do, stand in flat contradiction.

We found that it was as important to separate those two things as it bad been to distinguish logical from non-logical conduct, and our inductive survey showed that the failure to make such a distinction had been the main cause of error, from the scientific standpoint, in most social theories.

So we looked at them a little more closely and saw bow and why they went astray, and how and why, though fallacious, they enjoyed arid still enjoy such great prestige. In the course of that investigation we came upon things which we had not thought of at the outset. But we went on analyzing, distinguishing, and soon we observed another distinction that struck us as being quite as important as the others we had made--on the one hand an instinctive, non-logical element that was constant, on the other, a deductive element that was designed to explain, justify, demonstrate, the constant element. Arriving at that point, we found that induction had given us the elements of a theory.

Here, now, we are called upon to frame it, that is to say, we must now drop the inductive for the deductive method, and see what consequences result from the principles that we have found, or think we have found. After that we shall have to compare our inferences with the facts. If they fit, we shall keep our theory. If they fail to fit, we shall discard it.

In this chapter (and since the subject is a vast one, in the next two) we are to study the constant element a, going on, after that, to the deductive element b. But we are dealing with a very difficult matter, and a few more remarks in general on the elements a and b, and their resultant c, will not come amiss.

We saw in [an earlier section] that in the theories of the Iogico-experimental sciences one may discern a basic element A, and a deductive clement B, which in some respects are analogous to, in some respects different from, the elements a and b in theories that are not strictly logico-experimental.

The social sciences as hitherto cultivated show elements that bear a closer resemblance to a than to A, through their failure to avoid intrusions of sentiments, prejudices, creeds, or other predilections, tendencies, postulates, principles, that carry the thinker outside the logico-experimental domain.

The deductive element in the social sciences as hitherto cultivated sometimes comes very close to B, and there are cases where the logic is so adequate that coincidence with B would be exact were it not for a lack of definiteness in the premises a, which deprives the reasoning of strict validity. But oftentimes in the social sciences the deductive element stands very close to b, as containing many non-logical and non-experimental principles and showing great susceptibility to inclinations, bias, and the like.

So let us make the elements a and b our main concern. The element a corresponds, we may guess, to certain instincts of man, or more exactly, men, because a has no objective existence and differs in different individuals; and it is probably because of its correspondence to instincts that it is virtually constant in social phenomena. The element b represents the work of the mind in accounting for a. That is why b is much more variable, as reflecting the play of the imagination,

But if the element a corresponds to certain instincts, it is far from reflecting them all; and that is evident from the very manner in which we found it. We analyzed specimens of thinking on the look-out for a constant element. We may therefore have found only the instincts that underlay those reasonings. There was no chance of our meeting along that road instincts which were not so logicalized. Unaccounted for still would be simple appetites, tastes, inclinations, and in social relationships the very important class called "interests."

We may also have found only a part of one of the things a, the other part being a mere appetite. If the sex instinct tended only to unite the sexes it would not figure in our investigations. But that instinct is often enough logicalized and dissembled under guise of asceticism; there are people who preach virtue as a way of lingering, in their thoughts, on sex matters. Examining their thinking, we accordingly find an element a corresponding to the sex instinct, and an clement b that is the reasoning under which it hides. Diligent search might reveal similar elements corresponding to the appetites for food and drink. But in those cases the role played by simple instinct is far more considerable, at any rate, than in the case of sex.

The fact of being provident or improvident depends upon certain instincts, certain tastes, and from that point of view it would not figure in a. But in the United States the improvident instinct has fathered a theory that people ought to spend all they can earn; and so analysis of that theory yields a quantum a, which will be improvidence.

A politician is inspired to champion the theory of "solidarity" by an ambition to obtain money, power, distinctions. Analysis of that theory would reveal but scant trace of his motives, which are, after all, the motives of virtually all poIiticians, whether they preach white or black. First prominence would be held by principles a that are effective in influencing others. If the politician were to say, "Believe in 'solidarity' because if you do it means money for me," he would get many laughs and few votes. He therefore has to take his stand on principles that are acceptable to his prospective constituents.

If we stopped at that, it might seem that in the case before us the a's were located not in the principles that suggested championing the theory to the politician, but in the principles that inspired acceptance of it by his hearers. But going a little deeper, such a distinction is seen not to hold. Oftentimes the person who would persuade others begins by persuading himself and even if he is moved in the beginning by thoughts of personal advantage, he comes eventually to believe that his real interest is the welfare of others. Unbelieving apostles are rare and ineffective, but ubiquitous and ubiquitously effective is the apostle who believes, and he is the more effective, the more sincere his belief. The element a in a theory c is present both in the persons who accept and in the persons who propound it, but not to be overlooked in either case are the advantages accruing from the theory c, to the ones and the others.

In analyzing a theory c, we must keep the objective standpoint sharply distinguished from the subjective. The two researches are very often confused, and so two errors, in chief, arise. In the first place, as we have so often cautioned, the logico-experimental value of a theory is not kept distinct from its persuasive force or its social utility. Then again and this is a peculiarly modern error the obiective study of a theory is replaced by a subjective research as to how and why it was evolved or adopted by its author. This second research certainly has its importance, but it ought to supplement the other, not replace it. Whether a theorem of Euclid is true or false, and how and why he came to discover it, are two separate questions, and the one does not preclude the other. If the Principia of Newton had been written by an unknown writer, would that in any way affect the value of the book? So two of the aspects under which a writer's theory may be considered become confused, (1) his manner of thinking, his psychic state, and how he came by it(2) what he meant in a given passage. The first aspect, which is personal, subjective to him, is mixed in with the second, which is impersonal, objective. A factor in the confusion oftentimes is regard for the writer's authority. In deference to that sentiment it is assumed a priori that everything he thinks and believes must necessarily be "true," and that to determine his thought is tantamount to testing the "truth" (or when the logico-experimental sciences are concerned, the accord with experience) of what he thought.

Long prevalent was an inclination to consider theories exclusively from the standpoint of their intrinsic merit (sometimes their logico-experimental soundness), which, much more often, was determined with reference to the sentiments of the critic or to certain metaphysical or theological principles. Nowadays the tendency is to consider them exclusively from the extrinsic standpoint, as to the manner of their genesis, that is, and the reasons for their acceptance. Both methods, if used exclusively, are equally incomplete and to that extent erroneous.

The second error ( §855) is the opposite of the first. The first considered only the intrinsic merit of the theory; the second only its extrinsic merit. It appears in the abuse of the historical method, which is frequent enough nowadays, especially in the social and economic sciences. In the beginning, in their eagerness to free their science of contingencies of time and place, the fathers of political economy made the mistake of viewing their findings as absolutes. It was a salutary reaction, therefore, when just such contingencies came to be taken into account, and from that point of view the historical method was a notable contribution to the progress of science. And a forward step no less important was taken when the effort to derive the forms of social institutions from dogmatic absolutes was abandoned in favour of historical studies that made it possible to learn how institutions had developed, and their bearing on other social phenomena. We are altogether within the domain of logico-experimental science when we ask not what the family ought to be, but what it has actually been. But the historical study is to be thought of as supplementing, not as replacing, our inquiry into the relations between the constitution of the family and other social phenomena. It is useful to know how, historically, theories of income have been evolved but it is also useful to know the relations of such theories to the facts their logico-experimental value.

However, this latter type of research is much more difficult than the mere writing of history~ and there arc plenty of people who are utterly incapable of understanding, let alone of creating, a !ogico-experimental theory in political economy, yet who blithely presume to write histories of that science.

In the literary field historical studies often degenerate into mere collections of anecdotes that are easy to write and agreeable to read. To find out what a

writer ate and drank, how he slept, the clothes he wore, is intellectually and scientifically easier than to deal with the relations between his theories and experimental realities. And if a critic can find something to say about a writer's love affairs, he is certain to make a very entertaining book indeed.

To study the element b is to study the subjective element in a theory. But the subjective element may be further subdivided into two the general causes and the special causes that account for the genesis and success of a theory. General causes would be causes operative over fairly extensive periods of time and affecting considerable numbers of individuals. Special causes operate in an essentially contingent manner. If a theory comes into vogue because it serves the interests of a social class it has, in that fact, a general cause. If a writer invents a theory because he is paid to do so or because he wants to spite a rival, the cause is special.

Things that exert powerful effects upon the social order give rise to theories, and we shall find them, therefore, in the course of our quest for a's. In addition to such a's there are, as we have just seen, appetites and interests. Taking them all together we have the sum of the things that operate to any appreciable extent towards determining the social order (§851), bearing in mind of course that the social order reacts upon them, so that we are all along dealing not with a relationship of cause and effect, but with an interrelation or a relationship of interdependence. If we assume, as in fact seems probable, that animals have no theories, they cannot have an element a of any kind and perhaps not even interest all that is left in their case is instincts. Uncivilized peoples, however close to animals they may seem to stand, do have theories of one sort or another, and an element a has to be considered in dealing with them. And beyond a doubt they have instincts and interests. Civilized peoples have theories for very very many of their instincts and interests. An Element a figures through virtually the whole range of their social life.

In this volume we are to go looking for the element a. In many cases already we have distinguished a elements and b elements that we found combined and confused in some single phenomenon, c. That was in itself a start towards finding a norm for making such analyses. Suppose we get a still clearer view of the method from an example or two and then proceed with our systematic study.

Example 1. Christians have the custom of baptism. If one knew the Christian procedure only one would not know whether and how it could be analyzed. Moreover, we have an explanation of it: We are told that the rite of baptism is celebrated in order to remove original sin. That still is not enough. If we had no other facts of the same class to go by, we should find it difficult to isolate the elements in the complex phenomenon of baptism. But we do have other facts of that type. The pagans too had lustral water, and they used it for purposes of purification. If we stopped at that, we might associate the use of water with the fact of purification. But other cases of baptism show that the use of water is not a constant element. Blood may be used for purification, and other substances as well. Nor is that all; there are numbers of rites that effect the same result. In cases where taboos have been violated, certain rites remove the pollution that a person has incurred in one set of circumstances or another. So the circle of similar facts widens, and in the great variety of devices and in the many explanations that are given for their use the thing which remains constant is the feeling, the sentiment, that the integrity of an individual which has been altered by certain causes, real or imaginary, can be restored by certain rites. The given case, therefore, is made up of that constant element, a, and a variable element, b, the latter comprising the means that are used for restoring the individual's integrity and the reasonings by which the efficacy of the means is presumably explained. The human being has a vague feeling that water somehow cleanses moral as well as material pollutions. However, he does not, as a rule, justify his conduct in that manner. The explanation would be far too simple. So he goes looking for something more complicated, more pretentious, and readily finds what he is looking for.

The nucleus a, now that we have found it, is seen to be made up of a number of elements: first of all an instinct for combinations; people want "to do something about it"--they want to combine certain things with certain acts. It is a curious fact, also, that the ties so imagined persist in time. It would be easy enough to try some new combination every day. Instead there is one combination, fantastic though it be, that tends to prevail and sometimes does prevail over all competitors. Discernible, finally, is an instinct which inclines people to believe that certain combinations are suited to attaining certain objectives. 2

Example II. We |Lave seen many cases where people believed that they could raise or avert tempests. If we knew only one such case, we could make little or nothing of it. However, we know many cases and can identify a constant nucleus in them Ignoring, for the moment, the element in the nucleus that relates, as in the case of baptism, to the persistence of certain combinations and the faith in their efficacy, we find a constant element, a corresponding to the feeling, the sentiment, that a divinity exists and that, by a variable means, b, he (or "it") may be made to interfere and influence the weather. And then, right away, there is another sort of belief, the belief that it is possible to produce the desired effect by certain rites or practices, which mean nothing in themselves the practice, for instance, of tearing a white cock asunder and carrying the two halves around a field to protect it from drought. So the circle widens, and another constant a appears: an instinct for combinations, whereby things and acts designed for producing given effects are brought together haphazard.

Example IIl. Catholics believe that Friday is a day of evil <)men as so it is averred the day of the Passion. If we knew just that, and nothing else of the kind, it would be difficult to determine which of the two facts, the evil omen or the Passion, was the main, and which the secondary, fact. But we do have other facts of the kind, many of them. The Romans had their "black" or "vicious" days (dies atri or vitiosi), which were days of evil omen- for instance, the eighteenth of July, the anniversary of their defeat by the Gauls at Allia, A. U.C. 365. That is one kind of a--the feeling that the day which is associated with some catastrophe is a day of evil omen. But there are other facts. Both the Romans and the Greeks had days of evil omen and days of good omen without there being any special causes in the nature of public successes or disasters. Hence there has to be a more comprehensive class of a's, which includes the a just mentioned and expresses an impulse to combine days (and other things too) with good or evil omens.

These examples give us an inkling as to how a composite situation, c, may be broken up into a elements and b elements

Before going any farther it might perhaps be advisable to give word-names to the things we have been calling a, b, and c. To designate them by mere letters of the alphabet in a measure embarrasses our discussion and makes it harder to follow. For that reason, and for no other, suppose we call the things a, residues, the things b, derivations, and the things c, derivatives. But we must always and at all times remember that nothing, absolutely nothing, is to be inferred from the proper meanings of those words or their etymologies, that they mean respectively the things a, b, and c and nothing else.

As we have already seen, the residues a constitute a multifarious mass of facts, which have to be classified according to the mutual analogies they present. In that way we get "classes," "genera," and "specie." And so for the derivations B.

Residues correspond to certain instincts in human beings, and for that reason they are usually wanting in definiteness, in exact delimitation. That trait, indeed, nearly always serves to distinguish them from scientific facts or principles A, which otherwise bear some resemblance to them. Many times A's have come out of a's as a reult of making the a's more exact. The term "warm" is indefinite. Using it, it has been possible to say that well-water is "warm" in winter and "cold" in summer. But as used by physicists the term "warm" corresponds to certain degrees of heat as registered by a thermometer; it is definite. That made it evident that the water in wells is not in that sense warmer in winter than in summer, for a thermometer lowered into a well registers about the same temperature in winter as in summer, or if anything a lower one.

Curious the number of different meanings the term " warm " has in Macrobius, Saturnalia, VII, 6 8, all of them showing as their residue the sentiments that the term "warm" awakens in the minds now of this, now of that, individual. The doctors say that wine is warm; but a character in the Saturnalia disagrees, finding wine by nature cold. A woman's body, says another, contains a large amount of cold. No, answers a companion, the female body is naturally warmer than the male--it is so warm, in fact, that when is was the custom to dispose of dead bodies by cremation, a female corpse was commonly burned with each ten males so that the latter might more quickly be consumed. Women have so much heat in their bodies that they are able to wear light clothing in winter. Heat, moreover, is the principle of conception. All that is disputed by another, except as regards conception, the cause of which seems really to be heat. Why is it that in a very hot country wine has the property of cold instead of heat? The reason is that when the air is hot it drives the cold into the ground. The air is always hot in Egypt, so the cold permeates the soil and reaches the vine-roots, imparting its own properties to the wine. And we are told why a fan cools.

That is the type of the metaphysical reasoning, whether ancient or modern. The premises contain terms altogether devoid of exactness, and from the premises, as from mathematical axioms presumably trustworthy, conclusions are drawn by strict logic. They serve, after all, to probe not things but the notions that given individuals have of things.

The Macrobius example again shows how inexact terms may readily be used to prove both the pro and the contra. Women can wear lighter clothing than men because of the heat in their bodies. No, someone objects, it is because of the cold in their bodies. In general terms, it is the indefiniteness of the residues a, chiefly, that unsuits them to serve as premises in strict reasonings, whereas A propositions can be and are constantly being so used in the sciences.

The residues a must not be confused with the sentiments or instincts to which they correspond. The residues are the manifestations of sentiments and instincts just as the rising of the mercury in a thermometer is a manifestation of the rise in temperature. Only elliptically and for the sake of brevity do we say that residues, along with appetites, interests, etc. are the main factors in determining the social equilibrium, just as we say that water boils at 100 Centigrade. The completed statements would be: " The sentiments or instincts that correspond to residues, along with those corresponding to appetites, interests, etc., are the main factors in determining the social equilibrium." "Water boils when its calorific state attains the temperature of 100 as registered by a Centigrade thermometer. "

It is only by way of analysis and for the sole purposes of study that we distinguish various residues al, a2, a3 What is at work in the individual is sentiments corresponding to the groups (a1, a2, a3) ; (a_l, a₃, a₄); (a₃, a₅); and so on. These are composites as compared with the residues a1, a2... which are simpler. We might go on and break up a1 a2... as well into simpler elements; but we must know how to stop in time, because if made too general propositions end by meaning nothing. So the multifarious circumstances conditioning life on our globe may, in general, be reduced to solar light, the presence of an atmosphere, and so on; but the biologist needs conditions that are much less general than that as a basis for a greater number of biological laws.

It sometimes happens that a derivative, c, reached from a residue, a, by way of derivation, b, becomes in its turn the residue of other phenomena and itself subject to deviations. The bad omen, for instance, that is associated with the presence of thirteen persons at a table may be a derivative from a sentiment of horror at Judas's betrayal followed by his suicide; but that derivative has become a residue by this time, and people feel ill at ease at a table of thirteen without the least thought of Judas.

All the pointers just given must be kept in mind at all times in the investigations following. Anyone forgetting them will get everything askew.

This research as so far outlined has certain points of analogy with the ordinary researches of philology that deal with the roots and derivatives in which the words of a language originate. The analogy is not altogether artificial. It arises in the fact that products of the mental activity of the human being are involved in both cases, that their processes are the same. Take, for instance, Greek. The words in that language may be grouped in families, each family having its own root. There are the nouns meaning "anchor", "fish-hook", "curved object" "bent arm", "bend of the arm" "elbow" ; the adjectives "curved"and "hook-shaped"; the verbs "to fish with a hook" o) and "to bend" They all have the same root (residue), which originates in, and expresses, the rather vague notion of something curved, hooked, crooked. By processes of derivation, which have their rules, words are derived from these roots, just as the derivatives, c, are derived from the residues, a. We find combinations of roots just as we find combinations of residues. The adjective "biting a hook" has and for its roots, the first referring to something vaguely hook-shaped, the second to eating. There are some very common derivations in Greek. The suffix, for instance, combining with various roots, gives large numbers of words designating the effects of the actions indicated by the roots. So in social phenomena, certain derivations are very common. The Will of the Divinity, for instance, serves to justify no end of prescriptions. Combined with the residue of filial love, it yields the precept: " Honour thy father and the mother, for God so ordains. "

Actually observable in society are certain derivatives, c, that derive from residues, a, by way of derivations, b. Other derivatives may be as regularly deducible from the residues as the c's but are not observable in the concrete.

That situation has its philological counterpart in regular and irregular verbs. In point of fact such terms must not be taken literally. A so-called irregular verb is as regular as any other. The difference lies in the differing methods of derivation. A process of derivation used for certain roots gives a class of verbs that actually occur in the language. Used for other roots, it gives verbs that do not occur in the language. Conversely, the process of derivation used for these second roots yields verbs that occur in the language, but non-existent verbs when used for the other roots.

That is not all. The philologists of our time know that the language is an organism which has developed according to its own laws and is not an artificial invention. Only a relatively few technical terms, such as "oxygen," "meter," "thermometer," and the like, are products of logical activity on the part of scholars. Such terms would correspond to "logical actions" in society. The majority of the words in ordinary usage correspond in their formation to "non-logical" actions We have noted these analogies merely to facilitate a clear comprehension of the theories that we are expounding. They of course are not and could not be offered as proofs. Proof must come from direct examination of the facts and in no other way. The method that relies on analogies is a very bad method.

Investigations into the "origins" of social phenomena, which have so far concerned sociology in the main, have often times been, though their authors were not aware of the fact, searches for residues. It was taken for granted, more or less vaguely, that the simple must have preceded the complex- that the residue must have been anterior to the derivative. When Herbert Spencer locates the chronological origin of religion in the deification of human beings, he thinks he has found the residue of all religious phenomena, the simple phenomenon from which the complex religious observable in our day derive.

Two criticisms are to be made of that view. 1. No proof is offered of the hypothesis that knowledge of the residue is chronologically anterior to knowledge of the derivative. That has been the case in some instances, but certainly not in others. So in chemistry certain chemical compounds have been discovered later in time than the elements of which they are compounded, but many other compounds have been known earlier in time. In sociology the "latent" principles of law are an excellent example of derivatives that were known before their residues. An illiterate peasant woman in the mountains around Pistoia knows the conjugations of many Italian verbs by practice perfectly well and much better than any number of educated peoples but she has not the remotest idea of the rules that govern the derivation of those conjugations from their roots. 2. Even if knowledge of the residue is anterior in time to knowledge of the derivative, it is better to follow a course directly opposite to the one that has so far been followed. A chronological quest for the residue a is difficult, often impossible, because there are no documents for times so remote from ours; and it is illegitimate to take the imagination and the "common sense" of the modern man as substitutes for them. Imagination and common sense may, to be sure, yield fascinating theories, but they have little or nothing to do with the facts. To try to discover in primitive periods the residue, a, from which the phenomena, c, observable today, are derived is to try to explain the known by the unknown. To the precise contrary, the less well known must be inferred from the better known; one must try to discover the residues, a, in the phenomena, c, that are observable today and then see whether there are traces of a in documents of the past. If in so doing we find that a existed before c was known we might conclude that a is anterior in time to c, and that, in the particular case, the origin is one and the same with the residue. Where such proof is lacking no such identity can legitimately be assumed.

So far in these volumes we have tried, and we shall continue at all times trying, to explain facts of the past by other facts that we are able to observe in the present; and in any event, we shall always be at the greatest pains to work from the better known to the less known. We are not dealing with "origins" here, not because origins are not important historically, but because the question of origins has little or no bearing on the inquiry into the conditions determining the social equilibrium with which we are at present engaged. Of great moment, instead, are the instincts and sentiments that correspond to residues.

Chapter 9：Structural Functionalism