Positivism and Sociology: Explaining Social Life 
by Peter Halfpenny.
Allen and Unwin, 141 pp., £10.95, October 1982, 0 04 300084 3
Show More
The Rules of Sociological Method and Selected Texts on Sociology and its Method 
by Emile Durkheim and Steven Lukes, translated by W.D. Halls.
Macmillan, 264 pp., £15, November 1982, 0 333 28071 7
Show More
The Sociological Domain: The Durkheimians and the Founding of French Sociology 
edited by Philippe Besnard.
Cambridge, 296 pp., £24, March 1983, 0 521 23876 5
Show More
Durkheim and the Study of Suicide 
by Steve Taylor.
Macmillan, 249 pp., £15, July 1982, 0 333 28645 6
Show More
Show More

At a conference several years ago, a participant glared at the three sociologists who had just presented their papers. ‘The trouble with you is your rup,’ he said. Throughout the auditorium colleagues whispered to each other, ‘Rup?’ ‘What is rup?’ until the man on the platform translated: residual unresolved positivism. By accusing the speakers of rup, he implied that they might cleanse themselves of the remnants of positivism and become decent scholars again.

‘Positivist’ seems to have become the most demeaning name one can call a sociologist. It signifies a fetish for numbers (and for machines that grind and brew them) and an allergy toward social theory, philosophy, and anything abstract or otherwise unobservable, High in the ranks of those to blame for the residue of positivism in sociology has been Emile Durkheim, who among the founding sociologists (Marx, Weber, Simmel, Durkheim) was most sympathetic to positivist dogma. Unlike Marx, Durkheim is not represented by sections of sociological associations, nor do journals and political groups bear his name. Nor is his writing Germanically dense. Weber and Simmel experimented, almost playfully, with concepts and methodologies that might explain society, and their tastes in prose were toward the personal, historical and discursive – an invitation to generations of exegesis. But Durkheim fancied a science of society with rigorous inquiry akin to the biologists’ and findings stated as laws of cause and effect.

In Peter Halfpenny’s catalogue of 12 referents for ‘positivism’ as used by sociologists, Durkheim is ticketed for two: ‘the natural science of sociology consists of the collection and statistical analysis of quantitative data about society,’ and ‘science consists of a corpus of causal laws on the basis of which phenomena are explained and predicted.’ By my count, Durkheim can be charged with seven additional Halfpenny positivisms as well, since he held to the superiority of scientific knowledge, use of such knowledge to improve society, integration of various sciences, devoted study of society, verificationism, building facts systematically into laws, and tests of such laws.

Can Durkheim the multi-positivist be saved? Until recently very few sociologists have had reason to ask. On the one hand are those, including the whole of mainstream North American sociology, who go on about their research – using statistics and causal talk, scientificating by validation and the accumulation of findings – who seldom notice that they have been castigated as positivists, and who remember Durkheim as someone who a professor in graduate school said had started this line of work. On the other hand are qualitative researchers, including ethnographers, symbolic interactionists, ethno-methodologists, and followers of the latest French schools, who spice their scholarly reports with criticisms of Durkheim’s more evangelical claims about sociology as the supreme science.

A new impetus has emerged in the last ten years, however, with translations into English of several discerning essays by Durkheim – under the titles Emile Durkheim on Morality and Society and Sociology and Philosophy – which suggest the complexity and pertinence of his thought, and which have given rise to alterations in the received view of his theory. The obvious risk whenever this happens is of ending up with a theory of convenience, where one reads the texts in whatever ways seem most helpful to current projects or prejudices. Indeed several analyses just preceding this current lot failed in that way, consisting as they did of exhibits of passages where Durkheim sounds like the sort of phenomenologist or liberal the author wished him to be. The current reinterpretation differs, not only because it works with newly available data, but also because of its identification of a generative concept in Durkheim’s work: namely, homo duplex, a notion of man as both individual and social. As Durkheim explains in a recently translated essay:

Far from being simple, our inner life has something like a double centre of gravity. On the one hand is our individuality ... On the other is everything in us that expresses something other than ourselves. Not only are these two groups of states of consciousness different in their origins and their properties, but there is a true antagonism between them.

The notion of homo duplex turns out to be at work throughout all of Durkheim’s substantive research, displaying itself in his treatments of personal desires and societal constraints, individual and collective conscience, altruism and egoism, and the sacred and profane. In each instance, he proposes that a balance of the opposites is optimal and finds the fulcrum maintained by apparatus like laws and religions. Increases in social problems and deviance are manifestations of these embodiments of the ‘collective conscience’ becoming too powerful or too weak.

In quite a different way, the balancing of homo duplex is equally important for Durkheim’s methodology, because it prevents him from becoming the sort of social determinist who believes that all thoughts, feelings and social structures are produced by society, or a number cruncher who would reduce sociological phenomena to a set of correlations between variables, Which is why the new translation of Durkheim’s Rules is cause for celebration. Rules is his relentless call for a scientific sociology, and therefore provides an acid test for the current reading of Durkheim.

Through his translation, W.D. Halls has presented us with a Durkheim who is much less dogmatic and positivistic than that of the previous (1938) translation, and a relaxed and flowing syntax delivers Durkheim as an accomplished writer. Consider a passage in the earlier edition on the difference between social facts and facts about individuals:

But statistics furnish us with a means of isolating them. They are, in fact, represented with considerable exactness by the rates of births, marriages and suicides, that is, by the number obtained by dividing the average annual total of marriages, births and suicides, by the number of persons whose ages lie within the range in which marriages, births and suicides occur. Since each of these figures contains all the individual cases indiscriminately, the individual circumstances which may have had a share in the production of the phenomenon are neutralised and, consequently, do not contribute to its determination. The average, then, expresses a certain state of the group mind.

The same excerpt in the new translation:

But statistics affords us a means of isolating them. They are indeed not inaccurately represented by rates of births, marriages and suicides, that is by the result obtained after dividing the average annual total of marriages, births and voluntary homicides by the number of persons of an age to marry, produce children or commit suicide. Since each one of these statistics includes without distinction all individual cases, the individual circumstances which may have played some part in producing the phenomenon cancel each other out and consequently do not contribute to determining the nature of the phenomenon. What it expresses is a certain state of the collective mind.

The differences are substantial. In the new translation statistics ‘afford’ a means and are ‘not inaccurate’, and Durkheim distinguishes between suicide and what one can use to measure suicide. In the old translation statistics ‘furnish’ the means, offer ‘exactness’, and Durkheim talks of ‘the rates’. In addition, the new translation has Durkheim concerned with the ‘nature of the phenomenon’, the old with ‘determination’.

Or consider some sins for which Durkheim has been expatriated by anti-positivists. He treats social facts in thing-like terms, thus opening himself to charges of materialism. The old translation supports such a reading by talk of social facts exhibiting ‘extreme intangibility’ and coming into existence virtually of their own accord, while in the new translation these passages state explicitly that social facts reveal a ‘total lack of material substance’, and are brought about by the acts of persons. A related criticism concerns the old Durkheim’s image of the social world as a self-propelled system with causes ‘that are deterministic’ – a phrase the new translation omits completely. And some of the ten essays added at the back of this edition reveal that, at least in the later years of his life, Durkheim was somewhat structuralist about methodology. (Lévi-Strauss has indeed acknowledged that he works in the Durkheimian tradition.) In passages that obviate the normal work of contemporary positivist sociologists, Durkheim rejects generalisation as a criterion for acceptance of a finding about social phenomena, calling it ‘not necessary for the relationship to-recur more or less frequently; it is sufficient for it to be of a kind capable of recurring.’

Homo duplex also comes more fully into view in this new translation. Halls’s choice of words highlights the tension deriving from institutions which ‘bear down upon us’ but which we ‘cling to’ and ‘love’, as well as Durkheim’s emphasis on balance: ‘so that the originality of the idealist who dreams of transcending his era may display itself, that of the criminal, which falls short of the age, must also be possible.’ In some places Durkheim is almost majestic: ‘to provide a satisfactory explanation of social life we need to show how the phenomena which are its substance come together to place society in harmony with itself and the outside world.’

For all that, plenty of objectionable paragraphs remain in the new Rules – including pronouncements like ‘a legal rule is what it is and there are no two ways of perceiving it’ – and the biological metaphor dilated to such a degree that there are said to be species in the social world in much the same sense as among molluses and pigeons, complete with Durkheim’s famous postulate: ‘Each species has its own state of health, because it has an average type peculiar to it.’ Yet even here the book is germane to current debates in the limits it would place on sociobiology. Durkheim insisted that sociology ‘is not the appendage of any other science’, that ‘its subject-matter is an order of facts which other sciences do not study.’ The metaphors from evolutionary theory are merely that, and are not amenable, to direct application as by sociobiologists. The, central tenet of Rules (perhaps of the corpus of Durkheim’s work) is that things social exhibit patterns of development and structure which may parallel those biological and psychological, but are of ‘a nature sui genens’, because they emerge from ‘that special process of elaboration which individual consciousnesses undergo through their association with each other and whence evolves a new form of existence’.

A continuous dispute throughout the history of sociology concerns the appropriate data of this sui generis realm. When calling Durkheim a positivist, authors point to his calls in Rules for statistical analysis, and to his uncritical use of official death rates in Suicide. But Durkheim’s principal data in works like Division of Labour and Elementary Forms are ethnographic and historical – albeit collected by others – and in Rules he notes the value of comparative ethnography and is wary of the uncritical use of any data sources. Most important, Durkheim rejects the sort of work practised by contemporary positivist sociologists. Against their linear, ‘A causes B’ determinism, he views causes and effects as in a dualistic tension, wherein not only is the effect dependent but so also is the cause, since an effect ‘restores energy to the cause and consequently cannot disappear without the cause, being affected’. His most Durkheimian example is of punishment for crimes, a practice which derives from collective sentiments about right and wrong, these sentiments themselves being maintained and clarified through the public display of punishments.

The Sociological Domain reveals that these methodological issues were wisely debated among Durkheim and such followers as Bouglé, Mauss, Hubert, Simiand and Halbwachs. They argued about the possibility of isolating causes from effects of social facts, whether Durkheim was fetishistic in trying to copy the methods of the natural sciences, how to achieve explanations that balance the physical and mental aspects of the social world, and the proper uses of statistical and ethnographic data. These essays, taken together, indicate the special influence of Marcel Mauss, Durkheim’s nephew, on Durkheim and on the Durkheimians. Mauss brought ethnographic materials to Durkheim’s attention, and together they wrote an important monograph Mauss also became a kind of centre of the Durkheimian school, promoting the family creed, that sociology is the supreme science. Several authors in The Sociological Domain indicate, however, that this was mostly a tactical move on the part of Mauss and other Durkheimians. They were a group of former philosophers who framed their academic politics as a battle on behalf of science, and did so in an effort to make a place for sociology (and for themselves) in French universities. W. Paul Vogt sums the matter up: ‘the Durkheimians’ tendency to say they based themselves solely on the facts, while basing themselves mostly on good reaoning about the facts, led now and again to some embarrassing slips. Durkheim’s doctrine was, as Mauss pointed out, an uneasy mixture of rationalism and empiricism.

Steve Taylor, in an impressive first book, goes so far as to propose that sociologists who correlate ‘variables’ and ‘factors’ are, in Durkheim’s sense, non-sociological. As one concerned with the logic by which society grows and changes, Durkheim sought the formal structure within which concrete associations came to be played out. In Suicide, for instance, he wrote of polar societal currents such as egoism and altruism and anomie and fatalism, which when out of balance result in increased rates of suicide. Contemporary positivists operate without such a theoretical edifice and have thus heaped on one another thousands of correlations about suicide in the hope that eventually the shape of a theory will emerge from the rubble. Taylor spent several years reading through those studies and is an efficient rubbish collector in some parts of the book: he reveals fallacies in studies that correlate suicide with type of occupation, child-rearing techniques, business cycles, sex roles, mental disorders, religions. ‘While Durkheim was insistent,’ Taylor writes, ‘that society was something more than the sum of its parts, later students by and large abandoned this notion and concentrated instead on the relationships between the various and apparent parts of society.’ His purpose is to build an integrated sociological theory of suicide that rescues the good bits of correlational studies and is Durkheimian, not as a revisionist Suicide, but as a search for inherent and logical relations within society. While Durkheim made visible the functioning of groups – for example, that some over-integrate their members to the point of altruistic suicide for the sake of the group while others leave their members unaffiliated or desperately egoistic – Taylor carries on by looking at the act of suicide itself. He goes within the affiliations Durkheim set out between regulatory forces of society and rates of suicide, to identify relations between the manner in which people engage in killing themselves and the social circumstances of the persons involved.

To do this Taylor had to come to terms with the important missing component in Durkheim’s thought: what Steven Lukes, in his Introduction to the new Rules, calls ‘the problem of how to achieve and evaluate success in the interpretation of the actors’ world from within’. In his zeal to show that even the most individualistic phenomenon can be explained sociologically, Durkheims’s study of suicide concentrated on rates for religious and national groups and omitted individual persons, and research on suicide has since been conducted as a contest between explanations of individual actors and explanations of society.

Half of the book consists in Taylor’s elegant duel with both sides, and with the debates on positivist and interpretive methodologies. Faithful to Durkheimian homo duplex, he comes out holding for a balanced concert of subjectivist/meaning and external/societal explanations. He criticizes those who view suicide as resulting from the acts of persons involved, or worse, form the meaning these people hold, and the affiliated idealist position which takes suicide to be whatever those we study say it is. He is equally critical of most uses of suicide rates, which are biased toward over-counting some groups and under-counting others. Curious deaths by psychiatric patients or the unemployed are easily viewed as suicide, producing potentially spurious correlations between mental illness or unemployment and suicide. The statistics also ignore intentions of suicidal persons. Someone with little wish to die, who thought the car accident would not be fatal, is counted as a suicide when the ‘no one cares for me’ note is found in a pocket, but the person fully expecting to die, who jumped off the bridge without realizing a good swimmer was nearby, is not.

Taylor’s analysis relies less upon people as statistics than upon real and anguished men and women, whom he brings to us through the accounts of novelists and journalists, and directly through their autobiographies. In a powerful section, a woman whose husband has deserted her, learns she has cancer and writes a suicide note apologising to her daughter and asking her to stay ‘a pure and sweet girl always’; a man quietly kills himself in order to be with his wife in heaven, and the police officer says sincerely, ‘Lets’s hope he finds her’; and a psychiatrist obsessed with suicide is rescued after taking 45 sleeping tablets. He tell us about being reborn: ‘The over-intensity, the tiresome excess of sensitivity and self-consciousness, of arrogance and idealism, which came in adolescence and stayed on and on beyond their due time, like some visiting bore, had not survived the coma ... as the months passed I began gradually to stir into another style of life, less theoretical, less optimistic, less vulnerable. I was ready for insentient middle age.’

From these accounts – and from the study he has made of people found beneath Under-grounds trains – Taylor distinguishes between several types of suicide. There is the gamble with death where the outcome could go either way and the point of the act is to put oneself to the test in order to let fate decide whether one should or deserves to live, versus the submissive suicide, where one acts decisively to end life, and death is embraced because the persons feels defeated and wants out of the game. Both of these are characteristic of people at one pole of the Durkheimian schism: they are detached from others. On the other side are the famous ‘cries for help’ of those who commit suicide to seek love or attention from others. Taylor finds two distinct groups here too, according to the nature of the communication. Some use suicide to say that they want something from others or from situations; other utilize suicide to communicate something about themselves: for instance, feeling a burden on friends and family. The communicators are at Durkheim’s over-integrated pole. But Taylor corrects Durkheim’s theory in the light of data from suicidal persons themselves. Durkheim showed that societal imbalances in the direction of either collectivism or individualism result in significant increases in suicide rates, but left unattended a crucial question: why do only a small percentage of people in such circumstances actually commit suicide?

The question is one which positivist methodology pre-empts and which the qualitative methodologists have reduced to issues of meaning. From his struggles with these bodies of literature, Taylor concludes his book with a reply which he confesses is only in its formative stages. He suggests that attachment-detachment is only one of the ways in which homo duplex is played out in the case of suicide: it represents the dualism of the objective and societal level. Equally consequential – and complementary – is the subjective and personal level, where an interplay of certainty and uncertainty about existence is at work in the consciousness of individuals who engage in suicidal acts. For one group, suicide is a desperate game in which the outcome resolves the player’s uncertainly about whether it is better to live than to die, or is simply a sign that they are still alive in the midst of depression or physical suffering. For the other group, it is just the opposite. For them, suicide is an act of ultimate certainly. They are either at the end of hope and have found in death the only alternative still available to them, or else they are looking for something that life cannot offer.

Sociologists complain that the debate about positivism is interminable. A century of argument should have sorted out these matters sufficiently. Physicists don’t load up their journals and lectures with this stuff, nor do economists – a fact which students, colleagues and government ministers are fond of calling to out attention at embarrassing moments. But in a single year the haggling over positivism in sociology has provoked both a provocative study of one of the toughest substantive issues with which sociologists have concerned themselves, and an exciting edition of a classical text. Long live the debate.

Send Letters To:

The Editor
London Review of Books,
28 Little Russell Street
London, WC1A 2HN


Please include name, address, and a telephone number.

Read anywhere with the London Review of Books app, available now from the App Store for Apple devices, Google Play for Android devices and Amazon for your Kindle Fire.

Sign up to our newsletter

For highlights from the latest issue, our archive and the blog, as well as news, events and exclusive promotions.

Newsletter Preferences