[Updated in 2018 to just give you the article here
, but I’ll leave the rest just in case I forget that it took a few years to work out how to blog well.]
My new year’s resolution was to make a blog entry for every academic article published from 2013, since the article may be behind a paywall (although if you contact me, I will see you right) and the article’s ideas may be expressed in a relatively inaccessible way (although we don’t all spew jargon-filled group-closure nonsense). The aim is to get people interested enough to go from a short tweet to a larger blog to the high bar of reading (or the holy grail of citing) the article itself.
This article is called ‘Standing on the Shoulders of Giants
’ because I wanted to give the impression that we are discussing the accumulation of scientific knowledge; our aim is to build on the insights and knowledge produced by others rather than start from scratch each time. As stated, this is fairly uncontroversial and we might find that most people can get behind the project (in fact, they are already doing so, implicitly or explicitly). The more problematic and debatable part of this task relates to the details: *how* do we do it?
The article focuses on this task in the policy literature, but the themes extend to political, social and, in most cases, the so called ‘hard’ sciences. In fact, for many of us, it may be reminiscent of postgraduate discussions of the philosophy of science, in which we consider the inadequacy of most explanations of how knowledge is accumulated (from the ‘strawman’ of inductivism to the often-caricatured position of Popper (on falsification), to the idea of paradigm shift made famous by Kuhn and the rather-misleading ‘anything goes’ description of the approach by Feyerabend – a discussion captured neatly by Chalmers
). Many of us will have concluded two things: (1) we believe that we are in the business of accumulating knowledge/ we know much more about the world now than we did in the past, and we have acted accordingly; but, (2) we have no idea *how* that has happened because all of the explanations of knowledge accumulation are problematic, while some suggest that one body of knowledge *replaces* another rather than building on it.
In that broad context, the article (a) outlines three main ways in which scholars address this issue in policy studies and political science; and, (b) highlights the problems that may arise in each case:
1. Synthesis – we combine the insights of multiple theories, concepts or models to produce a single theory (in fact, the article discusses the difference between ‘synthesis’ and ‘super-synthesis’, but I don’t want to undermine my “we don’t all spew jargon-filled group-closure nonsense”). One key problem is that when we produce a synthetic theory, from a range of other theories or concepts, we have to assume that the component parts of this new hybrid are consistent with each other. Yet, if you scratch the surface of many concepts – such as ‘new institutionalism’ or ‘policy networks’ – you find all sorts of disagreement about the nature of the world, how our concepts relate to it and how we gather knowledge of it. There are also practical problems regarding our assumption that the authors of these concepts have the same thing in mind when they describe things like ‘punctuated equilibrium’. In other words, imagine that you have constructed a new theory based on the wisdom of five other people. Then, get those people in the same room and you will find that they will share all sorts of – often intractable – disagreements with each other. In that scenario, could you honestly state that your theory was based on accumulated knowledge?
2. The ‘Complementary’ Approach. In this case, you accept that people have these differences and so you accommodate them – you entertain a range of theories/ concepts and explore the extent to which they explain the same thing in different ways. This is a popular approach associated with people like Allison (who compared three different explanations of the Cuban missile crisis) and used by several others to compare policy events. One key problem with this approach is that it is difficult to do full justice to each theory. Most theories have associated methods which are labour intensive and costly, putting few in the position to make meaningful comparisons. Instead, the comparisons tend to be desktop exercises based on a case study and the authors’ ability to consider how each theory would explain it.
3. The ‘Contradictory’ Approach. In that context, another option is to encourage the independence of such theories. You watch as different research teams produce their own studies and you simply try to find some way to compare and combine their insights. Of course, it is impossible to entertain an infinite number of theories, so we also need some way to compare them; to select some and reject others. This is the approach that we may be most familiar with, since it involves coming up with a set of rules or criteria to make sure that each theory can be accepted (at least initially) by the scientific community. You may see such rules described as follows:
- A theory’s methods should be explained so that they can be replicated by others.
- Its concepts should be clearly defined, logically consistent, and give rise to empirically falsifiable hypotheses.
- Its propositions should be as general as possible.
- It should set out clearly what the causal processes are.
- It should be subject to empirical testing and revision.
For me, this is where the task becomes very interesting because, on the one hand, most of us will find these aims to be intuitively appealing – but, on the other, they are incredibly problematic for the following reasons:
- Few, if any, theories or research projects live up to these expectations.
- The principles give a misleading impression of most (social?) scientific research which is largely built on trust rather than constant replication by others.
- Many of the most famous proponents of this approach do something a bit different – such as when they subject their ‘secondary hypotheses’ to rigorous testing but insulate their ‘hard core’ from falsification.
- The study of complex phenomenon may not allow us to falsify, since we can interpret our findings in very different ways.
- Few theories are currently popular simply because they adhere to these principles. In fact, science is much more of a social enterprise than the principles suggest.
Of course, by now you may have identified a key problem with this argument: it is all beginning to sound a bit ‘postpositivist’ (which, in my mind, is still more of a term of abuse than ‘you, my friend, are a positivist’). However, it does not need to be taken this way. It is OK to highlight problems with scientific principles and admit that science is about the methods and beliefs accepted by a particular scientific community because, if you like, you can still assert that those principles and beliefs are *correct*. Many, many, people do. In fact, perhaps we all do it, because we have to find a way to accept some theories, approaches and evidence and reject others. We seek a way to produce some knowledge ourselves and find a common language and set of principles to make sure that we can compare our knowledge with the knowledge of others. We seek a way to sift through an almost infinite number of ‘signals’ from our environment, to pay attention to very few and ignore most. That task requires rules which are problematic but necessary.
All I suggest we do (which is a bit of a bland recommendation) is to reject the unthinking and too-rigid application of rules that hold us all up to a standard that no-one will meet. Rather, people in different disciplines might discuss and negotiate those rules with each other. This is more of an art than a science.
I also argue that (a) if we are serious about these rules, and the need to submit theories and evidence to rigorous testing; but (b) we accept that most of this is done on trust rather than replication; then (c) we should take on some of that burden ourselves by subjecting our own evidence to a form of testing, in which we consider the extent to which our findings can be interpreted in different, and equally plausible, ways. The article talks about producing different ‘narratives’ of the same evidence, but I won’t talk about that too much in case you confuse me with the presenter of Jackanory
Full reference: Cairney, P. (2013) ‘Standing on the Shoulders of Giants: How Do We Combine the Insights of Multiple Theories in Public Policy Studies?’ Policy Studies Journal, 41, 1, 1-21 PDF http://dx.doi.org/10.1111/psj.12000