The first paper in my second-year module Contemporary Social and Political Philosophy as part of my MA in Philosophy had to address the following question:
‘Functional explanations are essential for the study of evolution. Evolutionary explanation through chance variation and natural selection is a successful explanatory model. So it is surprising that functional explanation in the social sciences should be so contentious.’ Discuss
I am afraid I could not hide my deep disdain for functional explanations which I think don’t explain anything and which remind me more of conspiracy theories than of a scientific approach. So here is my paper:
In any debate – not only – in social or political philosophy, it won’t take long until someone mentions the word “society”. Society seems like a very real force. It seems to exist in the world just like other real things. But when we pause for a moment and look around, we only find human beings in this society, individual human beings. This leads some to go so far as to call society “fiction” or to postulate that “there is no such thing as society”1.
This debate is not only relevant for the normative aspects of social and political philosophy, but also for examining the explanatory approaches to social and political subject matters.
Functional explanation is one of these approaches, partly based on biology2 3 and especially the study of evolution4. This paper seeks to address the question whether functional explanation is an adequate explanation in the social sciences, especially in view of its origin in biology.
I. What is functional explanation in social sciences?
A functional explanation is one in which the existence of an entity or process is explained by the functions it carries out5 6 or by its effect 7, or put the other way round: the function of an institution, activity and so on is what explains its existence8 9.
To illustrate this with an example, the unequal distribution of wealth in a country may be explained by its function of keeping the capitalist system working, by allowing a small group of people who own most of the capital to employ the larger part of the population for wages which are just enough to survive but don’t allow the workers to build up a stock of capital of their own. If this sounds Marxist, it is no surprise because functional explanations are rampant in Marxist social science10 and actually dominated sociology until the 1960s11.
One characteristic feature of functionalism is that it postulates a purpose without a purposive actor13 14, just like evolution works without a purposive actor and is guided, if at all, by an “invisible hand“15.
II. Functional explanation in evolution
Evolution can be explained without the need for a purposive actor, as it works through chance variation and natural selection19 20. It roughly works like this: some animals of a species have genetic mutations which happen by chance. They have longer necks, stronger muscles, a different colour and so on, just like different human beings have different characteristics. Some of these mutations will be negative, they will make these animals slow or prone to disease. The animals affected in such a way will become extinct (possibly only over many generations) and with them the genetic mutations will die out. Other characteristics will prove to be beneficial by making the animals stronger, faster or give them better mimicry. These animals will live longer, find more mating partners, have more offspring and thus pass on their genetic sequences to far more animals in the next generation(s). This is how an originally arbitrary mutation will prevail over generations until a whole species will slowly have changed.
Pike describes this mechanism of chance variation and natural selection as leading to a “seemingly purposive result”21. I am not sure I would call it “purposive” because after all, it’s only chance. It is random. Some animals die, some species become extinct. What could be the purpose behind this?
I also find the description of evolution as “purposive” dangerously close to the theory of Jean-Baptiste Lamarck who did indeed argue that animals developed and acquired certain helpful features which were then passed on to the next generation. Since Charles Darwin, we know this theory of “soft inheritance” to be false.
III. Differences between evolution and social sciences
I have strong objections against the use of functionalism in the social sciences based on biological evolution, and I think there are numerous arguments against transferring this theory from the natural world to the world of social science, especially in highly complex societies. I shall address just a few of these arguments here:
- The main criticism of functional explanation is its lack of a mechanism22. Without a mechanism that is able to explain cause and effect, there is actually no cause and effect, but only correlation. Correlation can be an indication of causality in some cases, but doesn’t necessarily imply causality.
For evolution, we know this mechanism (but only since Charles Darwin): chance variation and natural selection. This is a logical mechanism, supported by ample evidence and meanwhile accepted by the overwhelming majority of natural scientists. – In the social sciences however, in the absence of a plausible mechanism, nothing is explained23.
Evolution is an inappropriate comparison to apply to social sciences because of the different time spans and the different subjects involved.
a. Life on earth began to evolve approximately 3.7 billion years ago. Mammals came into existence about 129 million years ago, modern humans only about 250,000 years ago.
These are obviously immense time spans over which no single scientist could observe much in the way of evolution. Even a scientist who would become very old and who would study one species all his lifetime could not determine the function or the purpose of a change in that species, if he could even detect any change at all. The “purpose” of an evolutionary change only becomes clear in hindsight.
In social science, we don’t have that much time. Changes happen much quicker, and we want answers much sooner. 23 years ago the world was divided in East and West, in a constant fear of a nuclear worldwide war. Just one and a half years ago we had dictatorships all over North Africa and no hope for change. Today, the world is completely different. Last year, during the London riots, a substantial part of society seemed to have changed within two days (and for not much longer).
A theory grounded in a science where developments can only be observed tens of thousands of years in hindsight is not suitable to social sciences.
Humans on the other hand are not only objects, but subjects. They think, they try to be rational, they have several choices, they make decisions, they correct these decisions, they talk to other humans about what to do jointly.
c. In addition to this rationality, humans are also blessed with (or plagued by) psychology. Evolution and the functional explanation of evolution doesn’t need to take account of psychology.
d. Humans – and other social actors like companies or government or armies – are much more different from each other than animals are.
A crocodile is a crocodile is a crocodile. At least for crocodiles of the same age, length and sex we pretty much assume that it will behave exactly the same way as any other crocodile with the same attributes.
Not so with humans: A 6-year old school girl in Kenya is completely different from one in Korea. A farmer in Iowa is completely different from one in Islamabad. Even if the economic and political situations were more or completely similar, people in different societies, countries or regions will still have different moral values, different priorities, different beliefs out of which they will act differently.
Evolution and the functional explanation derived from it is far too simplistic for the world as we know it.
Functional explanation is at best at the stage at which the study of evolution was at the time of Jean-Baptiste Lamarck and before Charles Darwin25: it may be true that a certain function is helpful in explaining the existence or development of a (social) phenomenon, but we cannot say for certain, we cannot say anything about the mechanism and we can neither prove nor disprove it (yet).
I will return to the very simple example of paragraph I., about the food in my fridge, to illustrate another point why I think that functional explanations are logically not sustainable. A functional explanation would explain the presence of the food by its purpose, i.e. to feed me. However, if I look into my fridge now, it is empty. Yet, I am hungry. What is the function of the absence of food or of the emptiness of the fridge? There is none that is discernible. Functional explanation cannot explain social phenomena that don’t have a function. But because functions are debatable – one person may see a function in something, the other may not – this leaves social science open to the kind of endless argument that is not worthy of the term “science”.
And the same applies to issues of more interest to social and political philosophers: if the function of the Arab Spring was to get rid of dictatorships, then surely this function would also have been fulfilled if the Arab Spring had happened two years earlier. Why didn’t it happen? The function, the goal, the effect would have been the same.
To me, this makes clear that there is no better explanatory method than methodological individualism which is based on the realisation that all actions are the actions of flesh-and-blood individuals and not of some overarching invisible hand27 28 29 and that rejects “all attempts to explain social … phenomena … unless they are couched wholly in terms of facts about individuals”30. As Karl Popper wrote: “All social phenomena … should always be understood as resulting from the decisions, actions, attitudes, etc. of human individuals, and … we should never be satisfied by an explanation in terms of so-called ‘collectives’.”31
Sticking to the example of the Arab Spring, some of the protesters may have been motivated by anger, some of them by poverty, some of them by ideas about liberal democracy, some of them by women’s rights, some of them just wanted to join their friends, some of them maybe craved the excitement and the action, some of them wanted to jump on the bandwagon in due time before the regimes fell. It may well be that there are almost as many motives and explanations as there were agents. But these explanations at the individual level can still be put together again to form a larger picture and then truly explain social change. This methodology will be more descriptive than functionalism, but it will stick to the facts and observations of researchable data instead of what will have to remain speculations about the function of a process which might end up having a completely different result once it will be completed in a few years, forcing us to change our view of the function altogether32.
Cuff, E.C., Sharrock W.W. And Francis, D.W. (2006) Perspectives in Sociology (5th edition), Abingdon, Routledge
Macionis, John J. and Plummer, Ken (2012) Sociology – A Global Introduction (5th edition), Harlow, Pearson
Matravers, Derek and Pike, Jon (2003) (editors) Debates in Contemporary Political Philosophy – An Anthology, Abingdon, Routledge
Cohen, G.A. (1982) “Reply to Elster on ‘Marxism, Functionalism and Game Theory”, Theory and Society 11 (reprinted in Matravers/Pike  and quoted by the pages of the reprint)
Elster, Jon (1982) “Marxism, Functionalism and Game Theory”, Theory and Society 11 (reprinted in Matravers/Pike  and quoted by the pages of the reprint)
Lukes, Steven (1973) Individualism, pp. 110-24 (reprinted as Methodological Individualism in Matravers/Pike  and quoted by the pages of the reprint)
Wright. E.O., Sober, E. and Levine, A. (1987) “Marxism and Methodological Individualism” New Left Review I/162 (reprinted in Matravers/Pike  and quoted by the pages of the reprint)
Pike, Jon (2005) The Philosophy of Social Explanation, A851 Issues in Contemporary Social and Political Philosophy, Chapter 1, pp. 15-26, Milton Keynes, The Open University
Margaret Thatcher Foundation, http://www.margaretthatcher.org/speeches/displaydocument.asp?docid=106689
The next paper, due on 4 May 2012, is about egalitarianism and sufficiency criteria which should be a bit more interesting.