ThoK
Notes

Social Sciences

  1. The Realm of the Social Sciences
  2. Are the Social Sciences Scientific?
  3. Explanations in the Social Sciences
  4. Case Study: Development Economics

    Green Readings

    Teaching Notes

 www.KAhome.eu 

Having defined knowledge as, roughly, 'true belief which can be appropriately justified' we have been asking all along what it is that counts as appropriate justification in different domains: the various areas in which we can have knowledge, such as morality and mathematics, all have their distinctive ways in which knowledge is gained.

In the case of the natural sciences, such as physics, biology and astronomy, we saw that knowledge is arrived at by the hypothetico-deductive method. In this section, now, we shall turn to the social sciences.

Exercise 0.1.:
  1. Based on what you have learnt before, try to describe the method of the natural sciences in a few sentences, using the following terms:
      auxiliary hypothesis,deductive reasoning,empirical,
      experiment,falsifiability,general laws,
      hypothesis,inductive reasoning,modelling,
      prediction,repeatability,testing.
  2. The natural sciences are said to be objective. What is meant by that, and what is their objectivity based on?

1. The Realm of the Social Sciences

The social sciences are often defined in terms of their subject matter, but there is no general agreement on what the academic disciplines are which constitute the social sciences.

Exercise 1.1.:
  1. What areas of study or research do we consider as social sciences?
  2. What do these various areas have in common, i.e. what defines the social sciences?
  3. What do the social sciences have in common with, and what distinguishes them from
    1. natural sciences, and
    2. literary criticism?
  4. Try to distinguish between a 'natural science' and a 'social science' approach in the treatment of patients with mental illness.
Exercise 1.2.:
A well-known law of physics states that there is an attractive force, gravitation, acting between any two masses.

By the next lesson, find one interesting theoretical statement of this general kind in one the social sciences, (which will be assigned to you.) You may use either your general knowledge, or look at books on the area in the library.

2. Are the Social Sciences Scientific ?

The social sciences are often criticised for not being 'really scientific', in the sense in which physics and chemistry clearly are. We will have to investigate in which ways they are different from the natural sciences, and whether this is a 'failure' on their part.

The issue, it seems to me, is muddled by the defensiveness of many social scientists: seeing the high -- usually unquestioned -- regard in which the natural sciences are held in modern society, they feel the need to justify their own, different work by denying any differences and appearing 'very scientific'.

In the first of the two brief quotations below, a famous economist explicitly compares his monetarist theory to a law of physics; in the second one, an American sociologist criticises the making of such comparisons. In the long passage that follows, another famous economist, writing more recently, explains where he thinks some of the problems lie.

As a scientific doctrine, the quantity theory of money (like the law of gravitation) is ideologically neutral.
Milton Friedman, The Times, 3.3.1980.
The possibility of a social science in principle as perfect as physics remains the unexamined premise of the vast majority of present-day social scientists.
Max Brodbeck.
Because the exchange of goods and services for money and each other is a less complex, perhaps less human activity than other social interactions, economics lends itself both to the development of theory and to the testing of that theory to a greater extent than other social sciences. If economics were a subject of purely intellectual interest like astronomy it would be regarded as a quietly progressive field, one in which there has been a steady accumulation of knowledge over the past two centuries.

But economics is not astronomy, because its conclusions have a direct impact on government policies that affect almost everyone. In an ideal world this would mean that large numbers of people would care about economics enough to study it closely. In our imperfect world it means that people care about economics only enough to know what they want to believe.

Politicization is, of course, not unique to economics or even to social science. ...

In the social sciences, it is much harder to draw this line [between serious ideas and pseudoscience]. Partly this is because one cannot perform controlled experiments: evidence in social science is always historical evidence, and history is complicated enough that its lessons are seldom unambiguous. Partly it is because social science studies people, and since we think we know ourselves, we all tend to think that we already know the answers.

The result is that while there is a steady accumulation of knowledge in economics, there is also a constant market for doctrines that play to popular prejudices, whether they make sense or not.

Paul Krugman, Peddling Prosperity, 1994.
Exercise 2.1.:
  1. Do you think that economic theory is value-free, or does it necessarily take an ideological or political position?
  2. Krugman also writes that ''we actually do know quite a lot about economics -- more than we know in any other social science -- because economics studies human beings in their simplest (if least edifying) activities.'' What activities is he talking about?
  3. Give examples of issues in the physical and biological sciences which have become politicized.
Exercise 2.2.:
In the natural sciences, certain requirements seem to be essential for us to gain knowledge. Discuss, with examples, how the social sciences fall short of satisfying these requirements.
Natural Sciences: Social Sciences:
a. Accurate measurements. There does not always seem to be agreement even on the observations on which the laws are based.
b.Repeatability. In many areas it is impossible to perform experiments, because the situations are not repeatable.
c.General laws:
''Whenever conditions C hold, event E will occur.''
The predictions made on the basis of the laws are often rather vague and open to interpretation -- the laws are 'bendy'.
d.Falsifiability. The observations are in many cases only statistical and descriptive of tendencies, and so cannot be used to refute theories.
Contrary evidence can always be explained away as an exception.
e.Objectivity. Theories in the social sciences seem to reflect personal opinions and prevailing ideologies, so they are not value free.
But as B.A. Farrell points out, there is a danger of confusion in criticising the social sciences for not being like the natural sciences. And in the long passage that follows, the economist Paul Krugman, who was already quoted above, not only defends his discipline, but sets the record straight on what can be achieved in the natural sciences.
We need only think of theories in fields such as history, economics, jurisprudence, social anthropology and sociology. These are (typically) not scientific narratives; but they are (typically) quite rational ones. The domain of the rational is wider than that of the scientific. Hence, to suppose that an unscientific theory is necessarily an irrational one is to perpetrate a howler.
B.A. Farrell, 1963.
An Indian-born economist once explained his personal theory of reincarnation to his graduate economics class. ''If you are a good economist, a virtuous economist,'' he said, ''you are reborn as a physicist. But if you are an evil, wicked economist, you are reborn as a sociologist.''

A sociologist might say that this quote shows what is wrong with economists: they want a subject that is fundamentally about human beings to have the mathematical certainty of the hard sciences. And without doubt there is too much mathematics in the economics journals, because mathematical elaboration is a time-honored way of dressing up a banal idea. But good economists know that the speaker was talking about something else entirely: the sheer difficulty of the subject. Economics is harder than physics; luckily it is not quite as hard as sociology.

Why is economics such a hard subject? Part of the answer has to do with complexity. The economy cannot be put in a box. Physics does very well at explaining simple, contained systems: planets orbiting the sun, electrons jumping between the orbits of a hydrogen atom. It has a much harder time when trying to cope with the complexities of nature in the wild: weather forecasting, even with massive expenditure on satellites and supercomputers, remains an inexact business. And when climate modelers are asked to answer a speculative question, like the prospects for global warming, they produce a range of answers (and a set of bitter disputes) as wide as that of economical forecasters asked to assess a policy initiative.

Another reason economics is hard is that the critical sociologist is right: it involves human beings, who do not behave in simple, mechanical ways. Economists understand hyperinflations ... about as well as meteorologists understood hurricanes before they were able to simulate them on supercomputers: that is, the principles and the rough magnitudes are well explained by the theory, even if it is hard to make precise forecasts. And a hurricane today is the same as a hurricane a thousand years ago; hyperinflations are something new under the sun ... And each hyperinflation is a little different from the last, partly because governments learn lessons (though not necessarily correct ones) from history.

Paul Krugman, Peddling Prosperity, 1994.
Exercise 2.3.:
  1. What do you think Farrell means by ''rationality''? Give examples of what then might be 'irrational activities'.
  2. Try to think of other questions in the natural sciences where ''forecasting ... remains an inexact business.''
In view of the 'suspicion' that the social sciences are unscientific then, we must ask whether we can know anything in the social sciences, and if so, what that knowledge consists of.

The answer is of course that while the social sciences to some extent follow the model of the natural sciences, they do have their own distinctive methods of explanation. To even compare their approach with that of the natural sciences is therefore in many cases quite inappropriate. In the next section we will be looking at some of the kinds of explanations characteristic of the social sciences.

Moreover, it might well be that making the social sciences as rigorous and 'scientific' as physics and chemistry, say, would require reducing their subject matter -- i.e. human beings -- to the level of physiology, of physics and chemistry. But is that what we want of the social sciences? Are the insights of psychology, for instance, of interest only until we are able to get a 'better', 'more scientific' understanding of events, in terms of neurons in the brain? (-- supposing that kind of reduction to be even theoretically attainable.)

The derivation of the laws of economics from the laws of physics and statements about which particles were where when might be in principle possible, but why want it? A few chapters of Keynes will tell one far more about regularities at the level of economic structure than such a deduction ever could.
Hilary Putnam, Philosophical Papers, II.

A Possible Conclusion

It seems that while the ways of attaining knowledge in the social sciences are not the same as those in the natural sciences, the two realms do share a broad approach that sets them apart from other fields. What follows is an attempt to account both for what the social and the natural sciences have in common, and for what differentiates them.

In the natural sciences, understanding a phenomenon consists of subsuming a particular case under a general law, and a scientific theory (typically employing some kind of model) often allows one to relate events as cause and effect. There is for example a well-known correlation between the height above sea-level and the temperature at which water boils:

increase in altitude above sea leveldecrease in boiling point of water
If we now ask why an increase in altitude results in a lowering of the boiling point, we will obtain a further explanation:
increase in altitude above sea leveldecrease in atmospheric pressuredecrease in boiling point of water
And if we ask, again, why an increase in altitude results in lower atmospheric pressure, we will again obtain a further explanation:
increase in altitude above sea levelless air above us attracted by the earth's gravitationdecrease in atmospheric pressuredecrease in boiling point of water
And so on.

Note that all the explanations here are of the same kind: each arrow links certain conditions to certain events, and represents a 'covering law' of the form: ''Whenever conditions C hold, event E will occur.'' Each why?-question can be answered, scientifically, only by an explanation referring to another, more basic law. (At some point, when we have reached the most fundamental laws, like the law of universal gravitation, there can be no further explanation -- except to say, non-scientifically, that it is because God made the world that way.)

Now, at first sight, the situation looks similar in the social sciences. There is for example the correlation observed in France around the turn of the century by Emile Durkheim, between instability, i.e. the rate at which society changed in an area -- be it for the better or for the worse --, and the suicide rate in that area:

increase in anomie ('lawlessness')increase in suicide rate
However, if we now ask why greater 'lawlessness' should result in a higher suicide rate, the explanation will not be in terms of further general laws; there is no direct causal link. Instead it will be in terms of the behaviour of individuals: when society objectively changes at a high rate, these changes affect the society's individuals, some of whom -- each for their own subjective reasons, such as disappointed expectations -- commit suicide, thereby contributing to the objective suicide rate. Similarly, although prices, rates of interest and of exchange, propensity to save, etc. are fundamental quantities in economic theories, the correlations which such theories claim between them are the result of summing up, over the whole of the economy, the effects of particular decisions by individuals.

The picture we get in a social science is therefore something like the following:

increase in anomie ('lawlessness')   increase in suicide rate
Having lost his traditional job and unable to adapt to a new one, Monsieur A. feels he is failing his family, so he drowns himself.
Madame B. cannot cope with her high expectations of her marriage being disappointed, so she takes poison.
And so on.

[ PS:  The above conception of the social sciences is not as original as I had thought; the same idea is expressed in what is known as the Coleman-Lindenberg diagram (Coleman, 1990) in rational choice theory:

 macro- (system-) causes Þ macro- (system-) outcome 
          ß   Ý          
  micro- (individual) causes   Þ   micro- (individual) actions  
]

And whatever sets the social sciences apart from the natural sciences -- such as that the evidence can often be only statistical; that the laws tend to be 'bendy' rather than rigid; that theories, and even the evidence, are not always value-free; that explanations are given in terms of 'ideal types'; etc. -- simply follows from the different nature of the relationships which underlie the observed correlations: causal relationships underpinned by a 'covering law' in one case, the behaviour of human beings in the other.

In some respects, however, they do not differ: the social sciences no less than the natural sciences are capable of being empirical and rational and objective, and of helping us to understand the phenomena around us.

Exercise 2.4.:
Try to give examples of actions, or changes in behaviour, by particular individuals which give rise to the following observations.
  1. The demand for most commodities is affected by their price: if the price goes up, the demand will fall.
  2. The relationship in a. does not hold for certain goods, and may even be inverted for luxury goods, such as perfumes.
  3. Less developed countries tend to have higher birthrates: when a country starts to prosper, the birthrate usually goes down.

3. Explanation in the Social Sciences

The social sciences encompass a range of approaches, and these bring with them different kinds of 'facts' and different modes of explanation of the facts.
In their treatments of the formal nature of knowledge, philosophers have distinguished between empiricism and rationalism. The former view sees knowledge as essentially a product of experience; the latter sees knowledge as essentially a product of thought. This distinction reappears in social science. There are those who think that social science is or ought to be like natural science, and thus essentially concerned with describing how things are in fact. And there are those who think that social science is a variety of social philosophy, and thus concerned with the interpretative understanding of patterns of culture. The adherents of both approaches make characteristic claims for the strengths of their views and vigorously criticize the other. The debates continue. [Two more passages from the book.]
P. W. Preston, Development Theory, 1996.
In this section we shall consider some of the different approaches and modes of explanation, starting with most narrowly 'scientific' one.

A. Behaviourism

There is a school of thinking which argues that the distinction between the social sciences and the natural sciences does not lie in their different kinds of explanations, but only in their subject matter: the concern of the social sciences is the behaviour of human beings. This view is taken by the school of behaviourism.

The school was founded by John B. Watson (1878 -1958) early in this century, and was a reaction against the prevailing 'mentalism', which was based on vague, unscientific introspection. In a paper in 1913, Watson boldly defined psychology as behaviourism, a ''purely objective branch of natural science'' whose goal was ''the prediction and control of behavior.'' Central to his rejection of mentalism was the claim that introspection was not a scientific method, because it rested on data known only to the observer. Over the next two decades, behaviourism became the central movement in psychology and other areas of the social sciences.

A major recent representative was B. F. Skinner (1904 -1990,) who investigated how we learn behaviour by reinforcement: he was consistent to the extent that he built a playpen for his own children based on the principles of operant conditioning. In the novel Walden Two, 1961, he envisaged a society in which everyone does 'what is right' because things are set up, and everyone has been trained, in such a way that what is right is just what one feels like doing; and in Beyond Freedom and Dignity, 1971, he argued that we should move towards such a society, and described the means -- some would call it mass conditioning -- by which it could be realised.

On the behaviourist approach, human actions are viewed exclusively from the outside: either in terms of stimuli and responses only -- like a mouse learning to find its way around a maze -- or in terms of inner states defined as states of the body and their characteristic behavioural expressions; (cf. Logical Positivism, Rudolf Carnap.) An example is the following passage, a behaviourist discussion of explanations of people's behaviour in terms of their attitudes:

The point I wish to make is that in imputing a certain schema of values to a community, one is imputing to its members certain attitudes. But an attitude is not something that can be established by introspection, whether in the case of our own person or of others. An attitude is a conditional or latent trait; and it is comparable in its theoretical status with viscosity or electrical resistance in physics, even if, unlike the latter, it can be usefully defined for socio-psychological purposes only in statistical terms. ... Whether ... explanations of social changes in terms of variations in attitude have a greater systematizing and predictive power than explanations employing different substantive concepts, is not the point at issue, and it cannot be settled by dogmatic a priori claims. But ... there is nothing in such explanations which differentiates them in principle from explanations in the natural sciences, or which requires for their validation a distinctive logic of enquiry.
Ernest Nagel, 1952.
Exercise 3.1.:
  1. How would you 'measure' an attitude, such as racial prejudice, and how the electrical resistance of a piece of material? In which way is such an attitude a property like electrical resistance, and not like a body's mass, say?
  2. Behaviourists claim their approach to be non-subjective, value-free, like the natural sciences. Do you agree with their claim, or are there particular choices involved in the behaviourist way of looking at human beings? (Consider the title of Skinner's book, Beyond Freedom and Dignity, for instance.)

B. 'Understanding'

In the natural sciences, the objects of study are detached from the scientist: there is no further answer than the law of gravity to the question why an object falls to the earth, it has no motive. So behaviourism, modelling its explanations on those of the natural sciences, also takes a detached view of human beings -- but thereby denies, or at least disregards, the fact that they may have motives for doing things, that they can have mental states.

There are areas of the social sciences, where a behaviourist explanation is sufficient: for instance when psychologists investigates the sensations of pain on skin, or the rate at which we learn and forget nonsense syllables. Moreover, when we study other animals -- and also kinds of behaviour that humans share with other animals --, a behavourist approach usually provides the appropriate level of explanation. But in many instances the behaviourist seems to be missing something essential about us as human beings.

The following passage eloquently criticises the behaviourist approach for over-simplifying matters, and therefore missing what is essential in our explanations of human behaviour:

From this point of view, the proponents of behaviourism in the social sciences fail to perceive the essential difference from the standpoint of causation, between a paper flying before the wind and a man flying from a pursuing crowd. The paper knows no fear and the wind no hate, but without fear and hate the man would not fly nor the crowd pursue. If we try to reduce it to its concomitants, we merely substitute the concomitants for the reality expressed as fear. We denude the word of meanings for the sake of a theory ... . We can interpret experience only at the level of experience.
R. M. MacIver, 1931.
In short, since the social sciences seek to establish 'meaningful' connections, and not merely relations of concomitance, its goals and methods must be fundamentally different from those of the natural sciences.

Exercise 3.2.:
What are the concomitants referred to above -- i.e. the observable consequences -- of fear, and of hate? Try to give definitions of these without referring to mental attitudes.

Exercise 3.3.:
  1. Consider the following scientific explanation:
    1. Whenever an object is not fixed and a force is applied to it, the object will move under the force;
    2. this paper has a certain surface area, and moving air exerts force on any surface;
    3. therefore the paper flies before the wind.
    This is a typical scientific explanation: from a general law and the particular conditions it deduces a prediction which can be tested.
    By imitating this explanation, attempt to formulate a behaviourist explanation of the man flying before the crowd.

  2. Discuss what is wrong with the following non-scientific explanation:
    The electron e- hated the negatively charged plate, N, and loved the positive one, P, so much that it ran away from N to be united with P.
Max Weber, writing on sociology at the turn of the century, had already introduced the distinction between causal explanations and understanding (in German: Verstehen); of these, the former could be said to be characteristic of the natural sciences, and the latter of the social sciences, (although Weber himself thought that all such explanations had eventually to be underpinned by a causal account.)
Human ... behaviour, like all events, shows both consistencies and regularities in its course. But what is specific to human behaviour, at least in the full sense, are consistencies and regularities whose occurrence can be interpreted to make it understandable. This 'understanding' resulting from an interpretation of some human behaviour has an immediate specific quality, which may be more or less strong, of obviousness [Evidenz]. That an interpretation is in this way particularly obvious, does not in itself prove its empirical validity. For one and the same action, identical in appearance and results, may be based on very different collections of motives, of which the most obvious and understandable one is not always the one that was involved. ... The highest degree of 'obviousness' is that of an interpretation in terms of rational purposefulness. ... [But] not only rational purposeful actions are understandable for us: we also 'understand' the typical course of affects and their consequences for behaviour.
Max Weber, 1913.
Exercise 3.4.:
  1. Contrast what Weber says in this passage with the behaviourist position.
  2. What do you think is meant by ''rational purposefulness''?
    Try to give examples of how we interpret actions in terms of their rational purpose.
While we can, according to Weber, '' 'understand' the typical course of affects and the consequences for behaviour,'' explanation in the social sciences typically goes beyond such empathising and assumes human behaviour to be rational: this is what is meant by the term ''rational man theory''.

This kind of explanation may seem 'unscientific', in that its approach is not so different from how we normally think in every-day life, and much of the terminology is taken from every-day language. However, the concepts and their use are rather more precise: what we normally call ''a family'' is not quite the same as what is meant by that term in sociology, for instance.

Moreover, while such explanations in terms of understanding may often resemble our every-day thinking, they do at the time go beyond it.

Because the application of rational man theory is an art, we can begin to understand why there should be room for a social psychology, side by side with common sense. Just by being possessed of common sense, we are equipped to practise the art of rational man explanation. But common sense, proverbially, is narrow. The justification of a rationalistic social psychology can only be that it provides a broader vision of the springs of action than that which people usually attain. The claim is that just like a Jane Austen, a George Eliot or a Henry James, someone like Goffman [a Canadian sociologist -- see below] expands ordinary horizons and lets us see at the origin of human behaviour concerns to which the habit of common sense normally blinds us. ...

But it is not enough to say merely that the explanations offered in social psychology are broader than those of common sense; some more specific mark of their alleged superiority needs to be named. ... If the social-psychological explanations really are superior hen it is natural to claim that their particular virtue lies in the fact that they create a greater potential for interaction.

A distinguishing mark of common sense accounts of actions is that when the ordinary patterns of accepted behaviour are broken, they resort quickly to what may be called quasi-rationalistic explanation. ... Typically, to suggest some examples, he [the agent] is taken to be overcome by passion, to be blinded by a physiological condition, to be the victim of some habits or illusions, or to be the product of a deprived or brutalising environment. By invoking such factors, quasi-rationalistic explanations have the effect of distancing us from the agents whose behaviour is explained: ...

Rationalistic social psychology, I suggest, is a significant discipline, because it often succeeds in providing full rational man explanations of human behaviour where common sense resorts to quasi-rationalistic ones. Thus, for example, it seeks to account for the actions of football hooligans or street vandals in a manner which reveals greater possibilities of interaction than the traditional 'They're just delinquent'. This latter story gives us no idea of how we might begin to have an exchange with the people in question, suggesting that the only possible reaction is the impersonal 'They need to be taught a good lesson' ... . A rationalistic social psychology would presumably take a different tack from this, since it would strive to relate the behaviour to beliefs and desires which, granted the circumstances of the agents, we find readily explicable. The account it offered might be less comforting for us, forcing on us the recognition that we are not different from them ... . Its virtue would be that it lets us see the agents as potential if not appealing interactants; however unpalatable the fact, it would allow us to recognise them as ordinary men.

Philip Pettit, ''Rational Man Theory'', 1978.
Exercise 3.5.:
Try to think of other examples in which a social science's rational man explanation of some behaviour goes beyond the quasi-rationalistic explanation one might give using only common sense.
(The prefix ''quasi-'' indicates that something appears to be something -- in this case rationalistic -- which it isn't really.)

C. 'Regular Tendencies'

Modern social scientists have generally given up looking for universal laws, and instead try to formulate regular tendencies. These may have a limited range of validity, culturally and historically; but they must be such that consequences can be derived from them that can be tested empirically, or checked against observations. Unlike in the natural sicences, though, these predictions will usually be probabilistic: that in certain situations, certain things are more likely to happen.

A famous example of such a theory is the study (mentioned above already) of suicide rates by the French sociologist Emile Durkheim, relating them to a ''lack of social regulation, or anomie'':

Society is not only something attracting the sentiments and activities of individuals with unequal force. It is also a power controlling them. There is a relation between the way this regulative action is performed and the social suicide rate. ... Every disturbance of equilibrium, even though it achieves greater comfort and a heightening of general vitality, is an impulse to voluntary death. Whenever serious readjustments take place in the social order, whether or not due to a sudden growth or to an unexpected catastrophe, men are more inclined to self-destruction.
Durkheim, 1897.
Exercise 3.6.:
  1. In the passage quoted no attempt is made to explain the correlation between the suicide rate and the state of de-regulation. Would you be looking for reasons or for causes?
  2. In the case of a particular person's suicide, when would you be looking for his or her reasons for it, for the reasons for it, or for the causes of it? (It might help to imagine different cases.)
  3. Suicide rates are fairly easy to observe. But how would you go about measuring anomie?
  4. Assuming ways of measuring suicide rates and anomie, how would you go about obtaining evidence for, or against, Durkheim's theory?
Even in the natural sciences, as David Hume first pointed out, and as we have seen before, causal relationships between events are not directly observable. All we can observe are ''constant conjunctions'', like between lightning and thunder. The causal relationships between the events are determined by 'covering laws' in our scientific theories, and as our theories change so will our ideas of what causes what -- as has happened with lightning and thunder.

And causal relationships are even more problematical in the social sciences, where we are very liable to two typical kinds of errors:

  1. 'Ecological fallacies' -- as when a high correlation is observed in a city between the rate of alcoholism in an area and the number of priests, (where there may be a third factor, of which both the observed ones are effects.)
  2. 'Pseudo-correlations' -- such as the relationship between the birthrate and the number of storks.

Exercise 3.7.:
Try to think of some other, more 'serious' examples of these two kinds of errors.

D. Ideal Types and Models

In the absence of formal theories and strict laws, explanations in the social sciences are often formulated in terms of simplifications or abstractions, which help us to understand essential aspects of a situation while omitting some of the specific details.

Thus, Max Weber introduced the idea of 'ideal types' (Idealtypen.) The term may be somewhat misleading, for there is no implication that something 'should' ideally exist or be a certain way. In the following passage Weber explains, (using history as an example,) what he means by an ideal type:

This conceptual pattern brings together certain relationships and events of historical life into a complex, which is conceived as an internally consistent system. ... In its conceptual purity, this mental construct [Gedankenbild] cannot be found anywhere in reality. It is a utopia. Historical research faces the task of determining in each individual case, the extent to which this ideal construct approximates to or diverges from reality -- to what extent for example, the economic structure of a certain city is to be classified as a 'city economy'.
Max Weber, 1904.
It would be pointless of course to redescribe every individual situation or institution, or any small set of similar situations or institutions, as an ideal type. An ideal type is well-chosen to the extent that it enables us to understand better the situations or institutions to which it applies, and to see correlations that we might not have seen without it.

Exercise 3.8.:
In many of his studies, the Canadian sociologist Erving Goffman defined particular ideal types, illustrated them with examples and investigated their common working. For instance, Asylums, 1961, is the study of the ideal type of a 'total institution': closed communities which seek to mould their inmates to some socially approved purpose.
  1. Can you think of instances of this ideal type?
  2. What kinds of behaviour might be characteristic of such institutions? Think of a variety of examples.
Models are similar simplifications or abstractions, now widely familiar from their use in economics. A model too tries to include the essential aspects, for the purpose of the present investigations, of a situation while abstracting from the non-essential ones. Examples of models are miniature train sets, and road maps, and the normal distribution: given just the mean and the standard deviation, the normal distribution enables us to predict quite accurately how many people in a population are taller than 180 cm, say, without having to measure everyone's height.

The use of models is not restricted to the social sciences, of course: we still imagine the atom as having a heavy nucleus with little electrons circling around it far away. But in the natural sciences, the purpose of such models is only to help us visualise the theory and get a feel for the underlying laws, they do not actually constitute the theory.

The question that one always needs to ask of any model is if it is a good model for what we want: a road map can be out of date, and the number of times that someone needs to toss a die until they get a Six is not normally distributed. The reason for using a model is that it is simpler to work with, the maths is manageable: so we can prove things in the model, which if the model applies well enough, will enable us to predict things about the real situation.

An example of the use of a model in economics is the multiplier effect: after introducing the 'marginal propensity to spend', we apply the mathematical formula for the sum of an infinite geometric progression. -- Here is another example:

In his celebrated paper, ''An exact consumption-loan model of interest with to without the social contrivance of money'' (1958), Samuelson addresses the following problem: Suppose individuals want to save for their old age, when they cannot produce anything, and suppose nothing last from one period to the next. All people can do if they don't want to starve is to strike a bargain during their working years so that those who are younger will support them with they are retired. In a world of endlessly overlapping generations of workers and retirees, what will the pattern of interest rates be?

To isolate the effect of this desire to provide for one's old age from other factors hat influence the rate of interest, Samuelson formulates an extremely simplified model in which everyone lives exactly three periods. In each of the first two periods of their lives, people produce one unit of a completely perishable consumption good , while in the third period of life, people produce nothing. To simplify further, Samuelson considers first an absolutely unchanging economy and then one growing eternally with an unchanging rate of growth and unchanging rate of interest. ...

Hausman, McPherson, Economic Analysis and Moral Philosophy, 1996.
Exercise 3.9.:
Try to use Samuelson's model: If there is no population growth at all, how much of what they produce would people have to save -- or rather: pay to those already retired -- in each of the first two, productive periods so as to maintain the same level of consumption throughout the three periods of their lives?

E. Functionalism

While causal explanations, both in the social and the natural sciences, are such that a present event (the effect) is explained in terms of past events (the causes,) there is another kind of explanation specific to the social sciences, where an event is explained in terms of its consequences: functionalism conceives of society (or an individual) as a self-regulating system which will try to maintain itself in some kind of equilibrium, usually without the members of the society (or the individual himself) being aware of that further purpose or meaning (the 'true function') of their actions (and institutions.)

One of the main representatives of this kind of thinking was the British social anthropologist A.R. Radcliffe-Brown (1881 -1955.)

The social life of the community is here defined as the functioning of the social structure. The function of any recurrent activity, such as the punishment of a crime, or a funeral ceremony, is the part it plays in the social life as a whole and therefore the contribution it makes to the maintenance of the structural continuity. ... Two important points where the analogy between organism and society break down must be noted. In an animal organism it is possible to observe the organic structure to some extent independently of its functioning. But in human society only the social structure as a whole can be observed in its functioning. ... The second point is that an animal organism does not, in the course of its life, change its structural type. ... On the other hand a society in the course of its history can and does change its structural type without any breach of continuity. ... There is not, and cannot be, any conflict between the functional hypothesis and the view that any culture, any social system, is the end-result of a unique series of historical accidents. ... One 'explanation' of the race-horse is to be found in its history -- how it came to be just what it is and where it is. Another and entirely independent 'explanation' is to show how the horse is a special exemplification of physiological laws. ... The two kinds of explanation do not conflict, but supplement one another.
A.R. Radcliffe-Brown, 1935.
Exercise 3.10.:
  1. Functionalism is often introduced by an analogy between society and an organism. Try to give a morphological description, and a functionalist and an 'historical' explanation of the stomach.

  2. Similarly, try to give a morphological description, and a functionalist and an 'historical' explanation of the institution of First-Year Induction.
    Are there, or can there be, aspects of the functionalist explanation which do not coincide with conscious intentions of the individuals involved?

  3. A functionalist explanation presumably must be capable of being founded on a step-by-step causal explanation. What feed-back mechanisms underlie your functional explanations of the stomach, and of First-Year Induction?

  4. Menenius Agrippa could be called an early functionalist when he addressed the rebellious slaves outside Rome in 494 B.C.:
    The prelude to the bloodless revolution was the First Secession of 494 B.C., when the plebs on returning from a campaign refused to enter Rome and withdrew to the Mons Sacer (the ''Mount of Curses'' rather than the ''Sacred Mount''). Menenius Agrippa at last persuaded them to return, showing by a parable of ''The Belly and the Limbs'' that they were a vital part of the state. [For a dramatic version of that parable, see Shakespeare, Coriolanus, Act I, Scene 1.]
    H. H. Scullard, A History of the Roman World from 753 to 146 B.C., 1961.
    What ideological doubts concerning functionalism does this give rise to? Consider also the following comment:
    Since it could be shown that differences in the work done (e.g. between a cobbler and a tailor) did not lead to differences in status ... The functionalists ... attempted to justify social stratification, (in particular the prestige associated with professional positions and the attendant privileges,) as functionally necessary. They argued from the different 'meaning' of the activities of the occupiers of those positions for society, and claimed that prestige and privilege are necessary to guarantee that important positions are filled.
    René König (ed.), Fischer Lexikon: Soziologie, 1958.

4. Case Study: Development Economics

The different views that can be taken of the social sciences and their role is not only a theoretical or ThoK-matter but has a direct bearing on policies, such as the approach that should be taken towards the countries of the Second and Third World, i.e. the countries of the former Eastern Bloc, and the less developed countries of 'the South'.

The traditional approach ... market- and structure-oriented ... interventionist ... later the New Right's free-market capitalism:

Positive economics is in principle independent of any particular ethical position ... Its task is to provide a system of generalisations that can be used to make correct predictions about the consequences of any change in circumstances. Its performance is to be judged by the precision, scope, and conformity with experience of the predictions it yields. In short, positive economics is, or can be, an 'objective' science, in precisely the same sense as any of the physical sciences.
Milton Friedman, Essays in Positive Economics, 1953.
Criticism of the traditional approach to development:
In sum, the following general points are made against the intervenors (shifting from knowledge, through expertise of ethic):
  1. that the strategy of descriptive-explanation, by way of making models, is epistemologically untenable, as the strategy misses precisely what is central to humankind, that is, its characteristic meaning-drenchedness;
  2. this being so, the claims that are made to a scientific status modelled on that of the natural sciences cannot be secured;
  3. relatedly it can be suggested that the repeated invocation of the status of science is in effect if not design obfuscatory;
  4. against the self-understanding of the practitioners it is asserted that what is in fact going on is the deployment of procedures of bureaucratic ordering;
  5. such strategies of bureaucratic ordering flow from the information needs and political agendas of dominant agencies of control;
  6. this being so, the key issue in respect of this material is not, as is claimed, empirical accuracy but instead political control; and
  7. the orientation of such work optimistically regarded is to the model of liberal-democracy, but more pessimistically such work is better read as concerned with absorption of other cultures within the expanding global capitalist system.

... finally, it is clear that the post-Second World War career of development theory can be read as a series of exercises in the construction and deployment of delimited-formal ideologies: arguments on behalf of those in authority and those with power.

P. W. Preston, Development Theory, 1996.
Hence, the alternative ... an agent-oriented approach:
The central claim made is that those involved must be seen as agents, as having their own understandings of their situation, their own expectations of change, and their own strategies for securing such objectives.

Overall, against the structural and interventionist orthodoxies of development theory three points are argued:

  1. that development studies must pay attention to the micro-scale detail of the social processes of the construction of patterns of life;
  2. that development studies must deconstruct the notion of intervention and shift away from untenable rational models of plan-making followed by plan-execution, and grant that intervention is itself a drawn-out and complex social process involving many agents; and
  3. that the further theoretical elucidation of these matters requires the supersession of the distinction between structure and agency via the detailed elucidation of these concepts, in particular the notion of agency. ...

In the light of such a view of the dynamics of the social world the familiar development theoretic concern with planned intervention comes to look very odd indeed. Out of the wealth of social interactions which constitute the social world the development orthodoxy is overwhelmingly concerned with one pattern of interaction, that of intervention, and this it construes in what upon examination turns out to be a deeply implausible fashion, seeing the business of intervention as one involving active and rational intervenors and passive, and maybe recalcitrant, recipients.

P. W. Preston, Development Theory, 1996.

'Green' Readings :

ECONOMICS
[1792; from Gk. oikonomos, household manager.] The exploration of wealth, value, and the distribution and management of resources. An important strand of green thinking about economics is Fritz Schumacher's ([Small is Beautiful,] 1973) concept of 'buddhist economics': ''the systematic study of how to attain given ends with the minimum means [and] ... the smallest possible effort.'' ... Part of the problems is that business executives have learnt the right language: ... [but d]o we take account of words or of action? ... How much of conventional economic theory and the existing economic establishment will be needed in a green society? ''Ever since the start of the Industrial Revolution, politicians and economists have worried themselves silly about supply and demand, productivity levels, the balance of payments, government spending or the international debt. ... It is no longer realistic to go out as we do now, ignoring basic human needs and deliberately destroying the planet in the process'' (Jonathan Porrit, 1984.)

COST-BENEFIT ANALYSIS
[1963] A technique evolved by economists which attempts to provide a 'rational' basis for decision-making by weighing up the value of the cost of each alternative against the value of the benefits thus gained. ... Economists frequently argue that using this technique to evaluate projects is at least better than using a technique that does not take both costs and benefits into account, but there are several important shortcomings to be considered. The most important criticism is that in order to be included in the calculation, both costs and benefits have to be measured, usually in money terms, which raises the question of value. During the London airport investigation, planners had to assess the 'value' of several medieval churches and a number of unique wildlife habitats -- an impossible task. A second criticism of cost-benefit analysis is that it appears to many people to provide a rational basis for decision-making, when in fact many of the 'facts' can be manipulated to provide the basis for any decision that those in power wish to take. Thirdly, and most importantly in many cases, cost-benefit analysis supposes that anything is negotiable, without seeing that some resources are finite and some costs completely unacceptable. For example, if a government proposes to base nuclear missiles in an area which contains the only known habitat for a particularly sensitive plant, the whole concept of the usefulness of cost-benefit analysis only makes sense if you believe that nuclear weapons have any value at all, and if you believe that a price can be put on the extinction of a species.

John Button, A Dictionary of Green Ideas, 1988.

The notion of 'perspective' adds a humanistic dimension to what otherwise may seem a rather academic way of choosing paradigms, namely by acknowledging 'human predicaments' as determinants of one's choice of objects of knowledge, methods and practical application. Different kinds of predicaments generate different perspectives. In the context of social science, as distinct from natural science, this humanistic dimension is inescapable whether or not it is made explicit. Once you take the point of view of a particular class of human predicaments, this provides you with a perspective and sensitizes you to particular problems, facts, methods of research and practice which are not recognized by somebody not sharing that perspective. ...

We often find the term 'predicament' used in literary essays, but less frequently in the treatises of social scientists, except perhaps implicitly in describing an exploited position in a class structure. ... We contend that it should be used more explicitly and systematically since the recognition or non-recognition of specific predicaments, and related action, would seem to constitute the very core of many controversies in the analysis of underdevelopment and development.

We believe that a 'predicament-free' analysis is closely associated with so-called 'value-free' science which, usually, is not value-free at all but is studying social and human reality from the vantage points of those who stand high above the predicaments, and who represent the hegemonous value-systems of ideological establishments, Western or indigenous. In this perspective 'from above' the main task of a social scientist seems to involve an explanation of causal mechanisms generating a particular state of affairs so as to enable a manipulation by 'social engineers' of relevant causes to attain certain valued effects. This is a paradigm of explanation, and a type of application, closely modelled on the shape of the natural sciences, and related technologies.

By contrast a predicament-oriented social science approach endeavours to attain two kinds of knowledge -- first a local understanding of the nature of given predicaments among those actually facing these predicaments in their everyday lives, ... and secondly a knowledge of the broader historical, structural and/or ecological causes generating such predicaments, [which in] the ideal case ... should be fed back to the local level to illuminate the understanding from below of the predicaments confronted there, and to provide guide-lines for local action and struggle.

Himmelstrand, Kinyanjui, Mburugu (eds.), African
Perspectives on Development,
1994. Introduction.

Teaching Notes :

1. The Realm of the Social Sciences
Exercise 1.2.: Each student should be assigned one of the following areas: Economics, Sociology, Geography, Anthropology, Psychology. Each of the subsequent lessons could then begin with
  1. a brief description, starting from student input, of what the object of study is of one of these social sciences, and
  2. a discussion of the statements students have found from that area.

The Sociological Perspective: In analysing and theorizing about the social aspects of human activity, sociologists have developed a number of important concepts. The most comprehensive of these concepts are the sociological definitions of society and culture. Sociologists use the term society to refer to all the social relations and groups formed by human beings; a society, as a singular unit, refers to members of a particular population occupying a particular territory.

The term culture refers to all ways of thinking, feeling, and acting that people learn from others as members of society (not only to so-called higher aesthetic activities). Culture includes nonsocial skills, such as how to plant crops, drive a car, play the violin, and so on. Apart from the social relationships in which they are taught and learned, such activities are not of primary concern to the sociologist; they are instead studied by the cultural anthropologist, who may take the entire culture of a society, from marriage customs to magic, as a field of inquiry.

Dennis H. Wrong, in Grolier's Academic Encyclopedia, CIS.
2. Are the Social Sciences Scientific ?
Exercise 2.1.:
  1. ''Creationists prefer not to believe in evolution, and they remain a powerful force well over a century after Darwin. A dangerously large number of people prefer not to believe that the HIV virus causes AIDS.'' -- other examples can easily be added.

3. Explanations in the Social Sciences
A. Behaviourism

Behaviorism is the movement and school of psychology that defines psychology as the science of the study of the behavior of humans and animals. Prominent behaviorists have included Edward C. Tolman, Clark Hull, and B.F. Skinner.

Prior to the founding of psychology as a science in the last quarter of the 19th century, it was considered part of philosophy and defined as the introspective study of the mind. The founders continued this view but introduced experimental methods. They also defined psychology more narrowly as the scientific study of consciousness, especially of its contents. They adhered to mentalism, the study of the mind.

Developments soon led psychologists to become unhappy with mentalism. For mentalists the central method of psychology was introspection, the precise description of conscious events. The notion of the human mind's uniqueness, however, was destroyed for scientists as they absorbed the meaning of evolution. Psychologists then ran into difficulty when they sought to investigate the minds of animals, which cannot introspect. The same difficulty confronted those who wished to study children or the insane. Lacking introspective reports from their subjects in these cases, they were limited to observing behavior and inferring mental events from behavior.

Especially important were the studies of the psychologist Edward L. Thorndike, whose pioneering studies of learning dispensed with introspection and reference to mind. In the 1890s psychologists developed mental tests, such as intelligence tests, that produced objective data about people without collecting introspective reports. Testing became especially popular in the United States, where psychologists were eager to develop psychology as an applied science capable of solving personal and social problems and serving the needs of business, industry, and education. They naturally focused on behavior instead of consciousness.

Finally, around the turn of the century, mentalism began experiencing controversies that to some young psychologists impugned its validity. For example, introspective psychology seemed unable to decide if all conscious experiences were sensations and images (memories of sensations), or if ''imageless thoughts'' also inhabited consciousness. One response was an American movement called functionalism, concerned with how the mind works rather than its contents. The functionalists were eclectic and used nonintrospective methods such as testing, but they remained mentalists who regarded psychology as the study of mind defined as conscious experience.

The existing dissatisfaction with mentalism was distilled in the founding manifesto of behaviorism, 'Psychology as the Behaviorists View It' by John B. Watson, published in the American journal Psychology Review in 1913. Watson was trained at the University of Chicago by the leader of functionalism, James Rowland Angell. While at graduate school, Watson became discontented with mentalism. He disliked serving as an introspective subject and felt that mentalism excluded, or at least looked down on, his own speciality of animal psychology. Leaving for John Hopkins University in 1905, he began to develop a new point of view, first for animal psychology and then, in the 1913 paper, for psychology as a whole. Watson boldly defined psychology as behaviorism, a 'purely objective branch of natural science' whose goal was 'the prediction and control of behavior.' Central to his indictment of mentalism was his claim that introspection was not a scientific method, because it rested on data known only to the observer. While disagreeing with his rhetorical excesses, most psychologists admitted that Watson's stance was correct. Over the next decade or two, behaviorism became the central movement in psychology.

From its inception, however, psychologists found it difficult to define behaviorism. Watson's manifesto was effective propaganda but poor conceptual analysis. Beyond defining the subject matter of psychology as behavior, consensus was elusive. Two problems were paramount. The first, less discussed, was the difficulty of defining behavior itself. To say it meant anything an organism did was far too broad. ''Molecular'' behaviorists wanted to define it as specific muscle movements, while ''molar'' behaviorists defined it as meaningful actions. The second problem was how to treat the traditional concepts of mind and consciousness. This was more openly discussed, and it led to the formation of different schools within the movement. The two most commonly identified are methodological behaviorism and radical behaviorism.

Methodological behaviorists were more numerous and included the leading theorists of the 1930s and 1940s, Tolman and Hull, whose followers were influential into the 1960s. While rejecting the scientific study of consciousness, they permitted reference to inner states of the organism, provided the states were properly defined. For example, the learning theorist Hull said that when an organism learns something it acquires a habit, an inner but nonconscious state inferred from behavior by the psychologist. Especially after the mid-1930s, methodologists were heavily influenced by the philosophical movement called Logical Positivism led by Rudolf Carnap. The positivists tried to define mental states as states of the body and their characteristic behavioral expressions.

Radical behaviorism was developed by B. F. Skinner. Influenced by the Positivism of Auguste Comte and Ernest Mach, he argued that exclusion of consciousness, or private events, from psychology was a mistake. It resulted in an incomplete science of behavior, since behavior is influenced by private stimuli such as headaches and private events such as conscious thoughts. At the same time, Skinner rejected the use of inferred organismic states. He held that private conscious stimuli and events were part of the natural world, even if observed by an audience of one, whereas inferred states were observed by no one, even their supposed possessor. Skinner regarded them as mythical. Radical behaviorism thus reverses the methodological behaviorist scheme, which excludes consciousness but includes inferred states.

A third school, strict behaviorism, was proposed by Watson's student Karl Lashley. Watson always emphasized the study of the nervous system, and Lashley -- influenced also by Ivan Pavlov -- developed this emphasis into a programmatic view of psychology. He argued that psychologists should explain both conscious events and behaviors as the result of processes in the nervous system. Thus, learning was to be explained by discovering how the brain and nervous system change as a result of experience. Lashley set out in search of the hypothetical unit of memory storage, the engram. Because of the limited technology of his time Lashley failed in his quest, and strict behaviorism remained a tiny movement.

In the 1960s a new movement developed called Cognitive Psychology. Disdainful of behaviorism, cognitive psychologists said they were returning to mentalism. In fact, cognitive psychology remains a form of methodological behaviorism because it studies behavior, not consciousness, and employs the strategy of inferring nonconscious mental processes from behavior. Radical behaviorism survived the death of Skinner and remains a healthy, if isolated movement within psychology. Lashley's strict behaviorism, without the name, is returning to psychology as modern techniques reveal the operations of the brain and nervous system in perception, learning, memory, and even thought.

Thomas H. Leahey, in Grolier's Academic Encyclopedia, CIS.
B. 'Understanding'

Exercise 3.4.:

  1. ''A rationally purposeful action shall be one which is in accordance, exclusively, with means (subjectively) believed to be adequate towards (subjectively) clearly conceived ends'' (Max Weber, 1913.)

C. 'Regular Tendencies'

Durkheim:

''So far is the increase in poverty from causing the increase in suicide that even fortunate crises, the effect of which is abruptly to enhance a country's prosperity, affect suicide like economic disasters.'' and:

''The limits are unknown between the possible and the impossible, what is just and what is unjust, legitimate claims and those which are immoderate. Consequently there is no restraint upon aspirations. ... Effort grows, just when it becomes less productive. How could the desire to live not be weakened under such conditions?''

D. Ideal Types and Models

Exercise 3.9.: ''In the case of no population growth at all, individuals can save one-third of their output in each of the two periods in which they are productive and then maintain the same level of consumption (2/3 unit) during retirement.''

E. Functionalism

Functionalism is a general social theory that stresses the mutual interdependence among the institutions and customs of any particular society. Functional analysis explains how social order is achieved by the functions that institutions -- such as the state, religious groups, the family, schools, and markets -- perform. For example, in complex societies like the United States, religion and the family support values that function to reinforce the operations of the democratic state and market economy. In simpler, tribal societies, participation in religious rituals functions to sustain social solidarity among groups related by kinship, but without centralized political institutions. Functionalist theory distinguishes between manifest and latent levels in the functioning of social institutions. In the preceding example, the manifest function of ritual, as understood by the tribe, may be to appease the gods, but its latent function, as understood by the social scientist, is to maintain group relationships.

Although basic to the writings of major 19th-century European social theorists, especially Emile Durkheim, functionalism was explicitly developed as a contemporary general theory by the sociologists Talcott Parsons and Robert Merton during the 1950s. This theory was the leading influence in Anglo-American sociology as well as in the other social sciences into the 1970s. Somewhat earlier than Parsons and Merton, Bronislaw Malinowski and A. R. Radcliffe-Brown, working separately, had developed a version of functionalist theory designed specifically for the research task of anthropology, focused on the study of small-scale, non-Western societies.

Since the 1970s functionalism has been modified to deal more adequately with the dynamics of social conflict. Functionalism is now cast in the sophisticated languages of systems theory.

George E. Marcus, in Grolier's Academic Encyclopedia, CIS.
Exercise 3.10.:
  1. The ideological reservation is that functionalism tends to defend -- or to be used as a defence of -- the status quo as necessary for the general good.

Concerning the need for a causal underpinning:

But the thrust of functionalism is towards explaining how actions not controlled in this deliberate fashion nonetheless contribute towards some socially useful goal. Thus, those unpublic-spirited characters, the old-fashioned political bosses, contributed to the integration of the immigrant into American society, without intending to do anything of the sort. However, in this case there does not seem to be much temptation to explain the behavior of bosses by reference to the beneficent effects they produce. The obvious answer to the question why Boss Tweed saw that the poor got coal at Christmas is that he wanted their votes. This indicates that where we are prepared to invoke final causes as explanations we want to be sure that the goal is not an accidental by-product of the behavior we are explaining; it is, of course, feed-back mechanisms which give us that assurance in the case of both organisms and self-regulating devices of the familiar kind. And it is their absence that makes such examples as the alleged 'latent functions' of bossism implausible.
Alan Ryan (ed.), The Philosophy of Social Explanation, 1973. OUP.
Reading :
Sue Bastian, ''Social Science''. Provisional Th.o.K. notes for teachers.

Ruth Benedict, Patterns of Culture, 1934. Houghton Mifflin Company, Boston, 1989.

Christopher Hookway, Philip Pettit (eds.), Action and Interpretation, 1978. Cambridge University Press.

Alexander Rosenberg, Philosophy of Social Science, 1988. Clarendon Press.

Alan Ryan (ed.), The Philosophy of Social Explanation, 1973. Oxford University Press.

Kenneth Thompson, Jeremy Tustall (eds.), Sociological Perspectives, 1971. Penguin. (Especially selections 5a and 9 by Weber, and 7a and 7b by Durkheim.)

Max Weber, Soziologie, Universalgeschichtliche Analysen, Politik, edited by Johannes Winckelmann, 1973. Kröner Verlag.

Bryan R. Wilson (ed.), Rationality, 1970. Blackwell.

Peter Winch, The Idea of a Social Science and its Relation to Philosophy, 1958. Routledge & Kegan Paul.