In The Rules of Sociological Method (1895), Émile Durkheim wrote down his own take on the (then) new science of sociology. According to Durkheim, sociologists like Spencer got it all wrong. The interesting thing about Durkheim is that he wasn’t just a criticizer, he actually offered an alternative approach. And in The Rules he sketches the outlines of the new science
.
The book itself is divided into six parts; each part deals with a specific problem for the new science of sociology.
In book 1, Durkheim explains what the object of study is supposed to be: social facts. But what is a social fact? A social fact is a thing that emerges from the interactions of individual people in a given society, which then introduces a restraining force on those very same people. Simply put, the way we think, feel and act shapes social facts, which then force us to think, feel and behave in certain ways. Social facts thus include things like family relationships, law, morality, religion, etc.
For Durkheim, all those things are collective representations, operating on the level of society as a whole, while manifesting themselves in individual lives. Sociology should study the things on the societal level, while keeping away from the individual manifestations. This means that induction is necessary: statistics and study of history and law show us the societal trends; the individual instances of these trends are un-scientific (at least, from a sociological point of view). So, in his earlier work The Division of Labour in Society (1893), Durkheim studied social solidarity as a social fact, by studying the differences in systems of law in different cultures, not by looking at individual manifestations of solidarity. And in his later work, Suicide (1897), he would use statistics to describe general trends in society which could manifest themselves in certain vulnerable peoples’ lives in case of suicide.
In book 2, Durkheim immediately answers how we should study these social facts. According to him, this should be done objectively, as opposed to the subjective approaches of his predecessors. The flaw in their thinking lies in their empiricism: they study social phenomena through their own preconceived ideas, which colour their experience. In other words, they study ideas and not facts. So, for example, economists studying the market mechanism of supply and demand study their own notion of it, not reality. Durkheim, in contrast, sees himself as a rationalist – someone who uses his reason to objectively study the facts by studying their indices. Studying social solidarity, for instance, requires observing the different systems of laws, as indices of social solidarity. One should leave his/her own ideas out of this research.
To be able to observe social facts, one needs to keep three rules in mind:
1. One must systematically discard all preconceptions.
2. One must define beforehand which group of phenomena are to be studied on the basis of common criteria, and all phenomena which correspond to this definition must be included.
3. One should consider observed social facts in isolation from their individual manifestations.
In short, this is a plea to end the subjectivism in social sciences, where the researches projects his/her own thoughts and feelings unto reality. At the time Durkheim was writing this book, this was a flaw that psychology suffered from as well – most psychologists tried to explain the human mind through introspection. But introspection doesn’t yield reliable and valid data; only by studying external, objective phenomena can one proceed in science. (Durkheim actually mentions the similarity with contemporary psychology.)
So far, so good. But in Book 3 Durkheim begins to mess up his own objective approach. In this book he deals with the problem of distinguishing the normal from the pathological. Durkheim seems to view himself as a diagnostician of society: through studying the state of affairs in a society, he is able to point to the pathological aspects of a given society and to point to their consequences. Of course, a diagnostician can’t sit on his hands when he has discovered pathology, so this immediately leads him to offer cures. Durkheim, for instance, sees socialism as a symptom of social pathology – something is going wrong in society (i.e. economic inequality and its corollary: power struggles) and many men suffer because of this social fact.
That this approach (distinguishing normal from pathological social facts) leads to controversial inroads, can be seen in Durkheim’s view on crime. He views crime as a normal social fact, a symptom of a healthy society, so to speak. Why? Because crime is inherently tied to humanity. But not only that, through enforcing the law on criminals (punishment), social solidarity on a societal level is strengthened. Yes, a terrible crime has been committed, but now we all see the law – our common law – in working and we feel all warm inside. We are one under the law – it’s just, sometimes we need a criminal to make us appreciate this.
(In my own view, Durkheim seems to mix up different types of ‘normal’ in his view on the pathological. There's statically normal, meaning a characteristic that falls within one standard deviation from the mean. Then there’s morally normal, meaning people who behave according to cultural expectations. Those who deviate are termed odd or transgressors. And of course, there are many more types of ‘normal’. For example, the religious ‘normal’ – not eating cows is normal for Hindu’s. It seems to me that Durkheim tries to smuggle in his own preconceived moral standard by judging what is normal and what is not.)
Next, in Book 4, Durkheim deals with social types. Since societies differ, to compare them means to define them in terms of different types. Comparative sociology stands or falls with the application of social types. While Durkheim observes an historical evolutionary trend in societies, he doesn’t think of social types as a linear progression from lower to higher, or simple to complex, etc. He likens the different types of societies (past and present) to a tree, much like Darwin came up with the ‘Tree of Life’ to illustrate the products of natural selection.
What are the criteria we use to class these different social types? According to Durkheim, these criteria need to be based on the social structure, or structural characteristics. Here, Durkheim seems to observe some sort of linearity: from fragmented small structures to huge, complex, concentrated structures. So, it seems that the development of administrative and political structures can be used to classify different types of society. This resembles Marx’ perspective on historical development, from tribes and clans to city states to kingdoms. The difference is Marx focused on the economic underpinnings, driving this historical development, while Durkheim focused on the structural characteristics of each society (politics, administration, demography, geography, etc.).
The fifth book is an important part of The Rules. In it, Durkheim sets out what a sociological explanation looks like. This might seem trivial, but it really isn’t. In science, an explanation is deemed to be valid if it establishes a causal connection between multiple phenomena. To establish the connection, one experiments with changing one variable while controlling all the rest. Differences in outcome are then deemed to be due to the different states of the variable. Sociology studies societies, the superstructure, so to speak, which makes experimentation practically impossible. (It shares this fate with all the other social sciences, economics included, and even cosmology.)
So if we cannot establish experimentally any causal relationship between social phenomena, how do we come up with reliable and valid sociological theories?
Here, Durkheim makes to important distinctions: (1) we have causes and we have functions. To describe a function of a thing is not causally explaining the thing. At best, the function of a thing can drive the needs or desires of individual people, setting in motion causal mechanisms. But the two aren’t the same. And then we still have (2) the difference between social and individual causation. For Durkheim, social facts emerge from individual behaviour and thinking, and social explanations should occupy themselves with explanations on the level of the social superstructure. Social facts can cause individuals to behave, think or feel in a certain way, just as individual behaviour, thoughts and feelings can cause social facts to exist or change. But the two should be separated clearly and sociology should occupy itself solely with social causation.
Durkheim’s view on causality is, unfortunately, very vague and ambiguous. It is also very outdated. For Durkheim a single cause is a necessary and sufficient condition for a single effect. Nowadays, statisticians and scientists work with multi-causal models, in which multiple causes create one effect, or a single cause creates multiple effects, or a combination of both. Also, nowadays we recognize multilevel causation, something which Durkheim steers clear from.
In short, Durkheim wants to explain social facts on the level of the society, in terms of other social facts. There is no recognition whatsoever of reciprocity between the macro-level and the micro-level (the individuals that make up such a society). It would be interesting how Durkheim would explain a bankrun in social terms. In the end, it’s individual people rousing suspicion and fear in the masses, who then en masse go the their bank and withdraw the savings. You can’t explain such a ‘social fact’ in terms of other social facts – somewhere along the line you have to open up your explanation for individual decisions on the micro-level.
In the last book (Book 6), Durkheim tries to mop up the mess he created in Books 4 and 5. He acknowledges the inductive problems that sociology is confronted with. To count as a science, sociology has to be able to demonstrate its hypotheses. And in abstracto this is done by ruling out alternative hypotheses. Normally experiments are used to rule out these alternatives; due to the complexity and fluidity of the subject matter, this path is closed to sociology. So how to demonstrate the validity of your sociological theories?
Durkheim argues that the sociologist is left with studying the different historical cases in their respective contexts. And when observing these cases, to use the ‘method of concomitant variation’. This means, basically, that the sociologist is looks for systematic variation, preferably serial variation, between different cases, and then drawing conclusions from these variations. System A differs systematically in x from system B; this happens also to be the case when comparing system B to system C, and system A to C; so the difference between systems A and B is due to x. So, for example, when you observe that suicide rates rise, and that these correlate with education levels, you look for systematic transformations in the given society. A secularization and the retreat of religion can then classify as cause for both effects: due to the retreat of religion the thirst for knowledge increases and the social bonds between non-affiliated people weakens.
But this is solely observing transformations in a given society, while for Durkheim sociology is almost identical with comparative sociology. He sees the study of different societies, past and present, as the only road to sociological knowledge. Comparing different societies of the same social types, preferable through time, offers us a wealth of information and might point us to social laws. Also, comparing different social types leads to unique insights, for example a genetic development of societies (how family ties develop in different social types). This last method will show us, according to Durkheim, the fundamental phenomena of sociology – we observe rudimentary social facts developing over time and between social types into complexes of “component elements” (p. 109).
As Steven Lukes explains in his fine introduction, the word elementary is crucial to understand what Durkheim is doing. It not only means elementary, in the sense of rudimentary, but also elemental, in the sense of the composing elements of a social fact. So in his last book, The Elementary Forms of Religious Life (1913), Durkheim sets out to describe how religion originated as well as what elements complex religions are composed of.
The Rules of Sociological Method is an interesting look at the attempts to raise a scientific artifice around the domain of social phenomena. Durkheim rightfully stressed the need to define the subjects of sociology (social facts); the need for definitions and objectivism; and the necessity of demonstrating the validity of theories. His main flaws (for which he can be forgiven, given that he was writing in 1895) are his mixing up ethics with science (normal/pathological) and his limited perspective on causality. The most important lesson to draw from this book – and this can be extended to psychology, economics, and like sciences – is that facts, not ideas, should be studied.
Just a quick glance at the contemporary humanities departments would make Durkheim turn over in his grave. One can laugh at his attempts, but studying solidarity through the type of laws or the pathology of society through suicide rates is a much more honest attempt than studying ‘black history’ or researching in ‘gender studies’. All these modern day social studies are ideologies dressed in the clothes of science.