There is virtually nothing in the physical universe around us that is not worthy of study. Most study begins with observations. We can term any such area of study where observations are made and knowledge accumulated as being a “discipline”. The social “sciences”, environmentalism and even astrology and palmistry could be considered disciplines.
But when does a discipline become a science?
I take a “science”” to be a field of study which pursues knowledge about the physical world around us in a particular manner. It starts by making observations. It could be in any field of study that the body of observations is eventually sufficient to allow the postulation of testable and falsifiable hypotheses; where the “scientific method” can be applied. The observations must be reliable and capable of being replicated. Theories about the mechanisms relating cause and effect can then be formulated. The assumptions for such theories must be explicit and clear. Based on the theories, predictions can be made for the purpose of testing the underlying theory or hypothesis. The predictions must be such that they are capable of being observed – whether by controlled experiment or in the natural world. Where necessary experiments and instruments and techniques can be devised to generate the observations which in turn must be repeatable and reproducible. While an incorrectly predicted result could prove the theory false, a correctly predicted result is not necessarily a test of falsifiability. The only valid conclusion from correct predictions is that a hypothesis may be true. The “proof” of a hypothesis or a theory comes not by a single decisive test of falsifiability but by building up a sufficient body of evidence of tests where the theory was falsifiable but could not be falsfied. Even where such a body of evidence has been accumulated it only provides “proof” that the theory applies within the bounds within which the assumptions of the theory apply. Extrapolation to regions or times where the assumptions of the theory do not apply cannot be valid.
The theory of gravitational attraction, for example, has been extensively tested, results can be predicted and it is today given the status of a Natural Law (where a Natural Law is one where compliance is guaranteed without intervention). Yet we still have no inkling of what causes this fundamental “weak” force of our Universe. It is a Law which seems to apply across all of the Universe as we know it but we cannot yet be absolutely sure that under some weird circumstances it may not apply!
With this definition where the application of scientific method is what distinguishes a science, it is apparent that there are many so-called “sciences” today which are – no doubt – academic disciplines but which are a long way from applying the scientific method. Foremost among these are Economics, Psychology, the Social “Sciences”, the political “sciences” and their off-shoots. (Note that Mathematics as a field of study is closer to being a language than to a science). But just as there are brand new branches of Physics and Chemistry where the observations are as yet insufficient to formulate theories and apply the scientific method, there are a few small branches of the “soft” sciences where postulated theories and even postulated “mechanisms” are beginning to emerge. It took a while for the discipline of alchemy to become the science of Chemistry. But it will take quite some time yet before Economics achieves the level of rigour needed to be called a science. These disciplines could be considered to be in a “pre-science” stage. Pre-scientific disciplines though should not be confused with “pseudosciences”.
In general usage, to be “scientific” is considered to be a “good” thing; objective, dispassionate, rational and logical. The downside is that it has become a label to be used to imply goodness where none is due. Similarly the ubiquitous label of being “scientifically proven” has become the stuff of advertisement campaigns where the scientific method and tests of falsifiability are nowhere to be seen. To be “scientific” or “scientifically proven” must – I think – be a conclusion to be reached after the evidence and cannot just be an assertion. Equally, to be “unscientific” is usually taken to be a bad thing even if the word only describes the absence of the application of the scientific method. In some fields which have not yet reached the stage of applying scientific method, there is nothing wrong in accumulating observations and even speculating on potential theories which are still not amenable to testing. That “scientific method” has not yet been applied does not disqualify the real observations that have been made. To merely dismiss such disciplines by referring to them as “unscientific” or “pseudoscience” is tempting but not justified – especially when they are pre-scientific disciplines. These terms are nearly always perjorative and are used only to denigrate. They may be useful in political or PR campaigns but they have no place in a scientific discourse.
But there are still far too many disciplines which do strive for academic and political approval as being “scientific” long before they can or are mature enough to apply the scientific method. Perhaps the term “pseudoscience” should be reserved only for these fields where political or other agendas lead to false claims of being a “science” or where a deliberate effort is made to appear to be “scientific”. Such disciplines tend to use a smattering of statistics as a “scientific” cloak under which to infer causal relationships without ever considering the mechanisms for the causal relationship. They tend to use a correlation as conclusive proof of a relationship rather than just as an indicator of a possible relationship and to initiate the formulation of a hypothesis. There are also many such fields of study which become fashionable by trumpeting their “multi-disciplinary” nature but while new disciplines can be created in this way it is not always true that the new disciplines have grown-up sufficiently to apply the scientific method or to be called a science. This is particularly evident in such areas such as “cognitive psychology” or “social anthropology” or “environmental sciences” or “ecological studies” or “the science of bio-diversity” or “climate science”. A mish-mash of legitimate scientific fields are mangled into mongrel forms and are passed of as “science”. Observations are often not repeatable. Data-sets are terminated prematurely or incorrectly continued to achieve the desired result. Conclusions are drawn in advance of evidence and then confirmation bias and cherry-picking of data are applied to “prove” the desired point. They tend to ignore the development of hypotheses and testing the falsifiability of these. Assertions from Authority and “consensus” of beliefs are taken to be proof of untested speculations. Instead of experiment and observation they often rely on the results of unproven and simplistic – but very complex and impressive-looking – mathematical models. Untested – but politically correct – model results take the place of real observations and are extrapolated – projected – into spaces and times where they cannot apply. Real observations which inconveniently contradict the model results are often discarded. Such “pseudosciences” are all too common but they are essentially a fraud.
But such frauds are not sustainable. They may well do damage while they remain in fashion but in the long run these “pseudosciences” will get “outed”. Astrology and palmistry are not taken too seriously these days but whether economic theories and climate “science” – which are still causing untold damage – are pre-scientific disciplines or are pseudosciences remains to be seen.
A valid discipline will – in time – mature into a science – if the evidence shows that it deserves to.
Tags: Causality, discipline, Falsifiability, Hypothesis, pseudoscience, Science, scientific method
The k2p blog 
November 12, 2012 at 7:46 pm
Clear, thoughtful, and has merit.
If the intent is to distinguish between something quite rare that deserves to be called “science” and other forms of inquiry, the entry is mostly not incorrect.
One shortcoming is that the actual practice even of particle physics is a lot messier than depicted: See Sharon Traweek on the maleness of physics and of cultural differences among nations. See Park Doing on the crucial role that technicians play, without whom the “scientists” could not do their work. See Latour and Woolgar on what actually goes on in laboratories.
The irreducibly social elements of scientific life put a different spin on the falsifiability ideas, but sociality can be made compatible with falsifiability. so it’s a smaller matter than the issue of whether falsifiability ought to be the gold standard for much of anything in human life. I doubt it.
As a social non-scientist, I observe that uncertainty and disagreement predominate in human life; and I believe that the purpose of organized inquiry ought to be to assist overwhelmed, confused, conflictual people and organizations to act more fairly and wisely. Falsifying in this author’s sense is rarely available for that task.
In contrast, it is fairly easy to cast doubt on beliefs that are widely held and grossly mistaken or biased. For example, the venerated U.S. constitution is typically not seen as setting up a system of government where multiple sources of veto (House committees, House floor, Senate committees, Presidency, courts, regulatory agencies) all can STOP or slow action. This means that there is little prospect of keeping up with highly dynamic technoscientific “progress.” Worse yet, the process for amending the constitution makes big change just about impossible.
I’m happy to agree that such an observation is not “science,” and hence that “Political Science” maybe should adopt a different title. And I acknowledge that my political observation is not beyond challenge — it just dents the fondest beliefs, but there are counter-arguments. Moreover, the example is among the softer issues on which organized inquiry might be helpful.
But if one took the redesign of toxic chemicals at the molecular level as a much “harder” case, my same basic point would apply. It’s impossible (in our era, at least) to know whether the synergistic effects of the 80,000 or so chemicals in commerce create very serious health dangers over and above those traceable to a handful of the most dangerous persistent organic pollutants. But everything from brain fog to fibromyalgia MIGHT be caused in part by the stew of toxics in our bloodstreams. Hence, anyone who can propose ways of making chemical greening more feasible could be quite helpful even if there are no “scientific laws” regarding, say, how to accelerate metabolism and excretion of toxics from living organisms. Cost may play a bigger role than chemical ideas in implementing Green Chemistry; but, say, the development of biocatalysis definitely is helping (though only a tiny bit of that area of inquiry may rise to the level of “science” as defined by our blogger).
Organized inquiry in my eyes, then, is a crucial adjunct and corrective to common sense — not a replacement for it. And aiming for falsifiability often is mistaken, because it takes away from doing what is needed and doable.
November 28, 2012 at 8:01 pm
[…] worthy of study and I would not go so far as to say it can never be a science but it certainly is no science yet. Stapel’s behaviour – which is not unique for publicity hunting social pseychologists […]
December 2, 2012 at 7:43 am
[…] behaviour is abnormal. Of course there is a vested financial interest for the discipline – which is a long way from being a science – to include as many behavioural conditions as possible within the definition of what is an […]