Lazy Re-run

Here are some thoughts on the philosophical differences between the physical and social sciences, because they may be of interest and they should at least help to clarify where I’m coming from.

I have a BSc in (Physical) Geography. It’s not exactly particle physics, but I spent my own fair share of time in the lab with the safety goggles on and it did give me a reasonable grounding in ‘the scientific method’. The essay below I submitted as part of my subsequent MA (in Environmental Politics, if you care), so there are probably a few copyright issues with pasting it up here. However, when I was digging it out I was slightly disappointed to find that it’s been almost a decade since I wrote it, so we should be fine. I’ve also long since lost the references file so anyone looking to rip if off for an essay mill will have to do at least some sort of work to get it up to standard. I also can’t remember the exact question it was meant to answer.

Anyway, here you go:

  

“The solution could have been reached immediately, if the questioners had but remembered that Hari SeldonHHhbhib was a social scientist, not a physical scientist and adjusted their thought processes accordingly...”

Isaac Asimov

Second Foundation1

Do the social sciences need to be more like the natural sciences? Asimov suggests that twenty-nine thousand years of galactic chaos depends upon the distinction between the two, so it would seem to be a worthy question. In this essay I will attempt to show that, while there are aspects of the natural sciences which social science should - and indeed need to - maintain, there is little reward in attempting to directly ape the predominant methods and philosophies of the natural sciences. There is perhaps greater potential in natural scientists moving towards social scientists than any potential cleaving between the two.

Questions about the methods and philosophies of the sciences necessarily involve questioning the aim of science. Is it simply to better humans’ lot in the world? Is it to fulfill some primeaval urge to understand? Or are there more spititual elements at work, where through science we hope to come closer to divinity? All these questions have been considered at great length elsewhere (Whitehead, 1925; Chalmers, 1999 among others), but crucially for the purposes of this discussion, the majority of these discussions have originated from the social sciences or philosopy.

Of course, such questions are not new, nor are they solely the preserve of social scientists, even allowing that “many of the traditional concerns of philosophy have been taken over by the social sciences” (Delanty and Strydom, 2003, p1). However, Rhoads and Thorn are not alone when they talk of a “basic sense of philosophical security” (1994, p 90) among physical scientists (in this case, geomorphologists). Where does this security come from, and why do social scientists not feel able to conduct their studies with the same security? Why, “in their hearts [do] social scientists deeply doubt the quality of their own explainations,” (Latour, 2000, p110)?

At the most fundamental level, natural and physical sciences are simply more tangible than social sciences. If you want to find out stress tolerances in a rock, you can hit it with a hammer. If you want to be more scientific about it you can take it to a laboratory and run a range of specific, established procedures on it to find more precise results. If you do this with a large enough sample of rocks then you can be pretty sure that, given a rock’s essential rocky nature, most other rocks of the same kind will perform in the same way.

This is the established mode of thinking in the physical sciences. Very well established indeed; in his Principia Mathematica (1687) Newton talks of the importance of this experimental methodology:

“For whatever is not deduced from the phenomena must be called a hypothesis; and hypotheses, whether metaphysical or physical, or based on occult qualities, or mechanical, have no place in experimental philosophy.”

A problem (for there are several) lies in the fact that most social sciences are not ‘experimental philosophy’. The procedure outlined above for hitting a rock with a hammer is problematic (on several levels) when applied to people and social systems. Social sciences are notably different from the natural sciences in that those studying are quite explicitly part of the objects and systems they purport to study; “the reflexivity of social science is to be attributed to the fact that social science is closer to the object of research than are the experimental and natural sciences, “ (Delanty and Strydom, 2003, p 2).

While it is true that physical scientists have needed to consider the way they interact with their objects of study (cf. Heisenberg’s Uncertainty Principle), many of these problems can be overcome through increasingly reductionist experimental techniques. Such soloutions are generally unavailable to social scientists; even if such experimentation were possible  there remains serious questions as to whether it is possible to draw broader conclusions in the same manner as for our rocks.

Physical scientists are aided by the fact that they are also essentially different from their objects of study. This difference allows for a clearer view of the subject. Social scientists again lack this difference and perspective, and how to overcome this is another crucial difference between social and natural sciences. One school of thought is to replace difference with distance; to try and remove the observer as far as possible from that which they are observing; but to “remove oneself at a distance from the object and then to surmount this very distance” (Bordieu, 1999, p395) introduces another set of complexities.

To briefly return to our opening quote, Hari Seldon was the greatest social scientist who ever (will have) lived, and his plan depended on the secrecy of his sect of social scientists (the eponymous Second Foundation); which if anything suggests similar thought processes to physical scientists, not a galaxy changing difference. If the wider population ever knew of their existence the plan would be ruined. For many years social scientists in the more prosaic world tried to uphold similar standards of seperation through the impartial authorial voice of logical positivism or critical realism.

However this impartiality is at best a chimera. There are many other reasons why such impartiality is impossible in the social sciences (see Haraway, 1988; Cronon, 1992 among many others) and the reality (as such) is closer to what Haraway calls the ‘partial perspective’.

Given then that the social sciences are at once both part of their field of study and capable of understanding only a part of it, there is limited scope for effectively adopting the kind of reductionist experimental methodologies of the natural sciences. Indeed; “the imitation of the natural sciences by the social sciences has so far been a comedy of errors” (Latour, 2000. p114).

That is not to say though that there is no part of the methodologies or philosophies of the natural sciences which could not be usefully adpoted by the social sciences. Given that the basic ‘experimental philosophy’ of the natural sciences is (essentially) inapplicable to the social sciences, what should take its place? Is it in fact necessary for anything at all to take its place?

Yes, it is. Despite arguments for an ‘anything goes’ approach to method in the social sciences (Feyerabend, 1975), to follow such an approach would risk invalidating much of the work done by social sciences. Given that the social sciences directly interact with their object of study (‘society’, for want of a better term) in a way the natural sciences do not; that “social sciences create phenomena through the procedures that are established to discover them,” (Osborne and Rose, 1999, p367) it is necessary to pay close attention to the realtionship between the social sciences and society.

Society has a confused relationship with science, and appears to make no distinction between social and natural scientists. Both are labelled ‘Boffins’ by the tabloid press. Indeed, as part of society this confusion extends to academia itself. Some of the suggested methods for social science essentially amount to literary criticism for the analysis of a single text (e.g. Franzosi, 1998), which may hold great clues for understanding the author, but limited potential for wider society. How then does social science move away from literary criticism and ad hoc improvisation towards a more ‘scientific’ way of doing things?

Science is essentially about increasing understanding, and the main problems with an ‘anything goes’ approach is that it relies upon the inspiration of individuals. There is limited room for Newton’s ‘standing on the shoulders of giants’ if a flash of inspiration is inexplicable to others. It is misguided to argue that the existence of some kind of method rules out any such inspiration, even given that “there is no general account of science and scientific method to be had that applies to all sciences at all historical stages in their development” (Chalmers, 1999, p247). Epistemic shifts do occur², methodologies change, but that is no reason to cast them off altogether.

The important aspect of a methodology with regard to the social sciences is less about what the methodology itself is, and more about rendering research and findings transparent to others. Without such transparancy it is all to easy to level accusations of, at best incoherence, and at worst quackery or mysticism. If the results and their applications are incomprehensible to the wider community (whether that community is academic or otherwise) then the opportunity to progress is annulled. Feyerabend may dismiss observational reports as ‘magical phrases’ (1975), but the existence of some kind of repeatable, verifiable methodology (be it observational or otherwise) is one of the fundamental differences between science and the mystical realms of astrology³, or indeed science fiction.

Methodology is perhaps the wrong word to use in regard to social science, however. Both Cronon (1992) and Baxter and Eyles (1997) use the term ‘rigour’ to describe an explicable methodology, as “for the research to be evaluated, there must be clarity of design and transparency in the derivation of findings” (ibid, p506). This ‘transparency of derivation’ is essential, it is perhaps the defining characteristic of all kinds of science. For this reason it is important for social scientists to avoid the trap of “try[ing] to follow natural scientists as closely, or as far away, as possible” (Latour, 2000, p114).

The claim that social science needs to become more like natural science is therefore unhelpful, for social science must necessarily follow different research paths and methodologies from the natural sciences. However, as a science, of whatever hue or orientation, certain demands are made which must be fulfilled. In this respect both natural and social sciences share common ground.

It is possible to counter the claim with its opposite; that the natural sciences need to become more like to social sciences. While pyhsical scientists may feel that their difference from their subjects of study provides them with a certain philosophical surety, “the philosophy of the physical sciences... is not as secure or uncontroversial as it seems” (Rhoads and Thorn, 1994).

There are (at least) two ways of looking at this claim. The first is readily recognisable from the discussions about production of knowledge by the social sciences, that;

“the production of scientific knowledge always takes place in a social context in which that aim is interrelated with other practices with different aims, such as those involving the personal or professional aims of scientists, the economic aims of funding agencies, the ideological interests of religious or political interests of various kinds and so on.”

Chalmers (1999), p248-9.

It could well be argued that considering these aspect of scientific enquiry falls within the sphere of social scientists. However the second aspect is less easy to qualify as a social science issue, and that is problems arising (mainly) from the relativist schools of thought regarding the ‘truth’ of scientific enquiry. To return to our exemplary rock; initially I outlined a mode of natural scientific enquiry founded on certain empircal certainties, one of which being of course that the rock exists⁴. It would be possible to end doubt about the rock’s existence by physically throwing it at the head of anyone who disagrees, and on such certainties are the physical sciences based. Less certain however is the assumption/induction that the traits shared by a large number of rocks will be shared by most, if not all, other rocks of its type.

Real problems start to occur when we move beyond the world of objects we can manipulate with our hands. When we venture into the world of the very small or the very large we necessarily have to work, not with tangible, tractable rocks, but with metaphors. Metaphors, as with all linguistic constructions, are only ever approximations of real events and objects, and subject to revision or abadonment as understanding increases; which leads to the conclusion that, “science does dot progress towards towards truth; it merely changes its communal perspective of reality” (Rhoads and Thorn, 1994, p94).

What does it matter though, if the results of these metaphors is to allow us to reliably fly halfway round the world in less than a day (Chalmers, 1999)? It matters because, just as a lack of ‘rigour’ in social sciences can hinder wider understanding, so too a lack of philisophical consideration can remove an important tool for understanding in natural sciences. If there is no absolute ‘truth’ can there be absolute laws? As Newton’s laws of motion have been superseded by general relativity, is it possible that all such laws may be revised not by more fundamental laws, but by better metaphors?

Just as there is no outstanding ‘need’ for social sciences to become more like the natural sciences, so the reverse is also true. That said, to continually insist on some kind of fundamental dichotomy between the two actively hampers progress in both, not least because some of the most major impacts and shifts in human history have happened at the intersection of the two.

---

1 (1964) Emphasis in original.

2 Delanty and Styrdom (2003) identify four such shifts in the social sciences in the twentieth century.

3 Clarke’s third law states that ‘any sufficiently advanced technology is indistinguishable from magic’ (Clarke, 1973) and highlights the importance of research being understandable and explicable.

4 Which, of course, it doesn’t.