Is psychology a science?

Periodically one encounters the question of whether this or that topic or discipline is a science or not a science. This is a venerable problem that even has its own name (the demarcation problem) that I have written about extensively in the past (see my 2011 series of posts on the Logic of Science) and the consensus has been that it is impossible to specify both necessary and sufficient conditions that are necessary to do so. In most cases this inability to construct a strict demarcation rule does not really matter in any tangible way. After all, what does it matter what label you give something? But unfortunately it is the case that being considered ‘scientific’ adds a certain authority to statements, which is why people invoke it so frequently.

But even one can’t get complete demarcation criteria, one can usually specify just necessary conditions (such as being naturalistic, empirical, testable, predictive) that can at least identify what is not science if they do not meet those minimum criteria. It is just this question that is usually important anyway. So, we can say that creationism or intelligent design is not science while we cannot say that string theory is not science. In ambiguous cases, it is the collective judgment of the community of scientists that is the ultimate arbiter.

It is similar to art. I would argue that it is impossible to create demarcation criteria that determine what is art and what is not art, but we leave it up to the experts in the artistic community to make the final judgment, even if we may not agree with some of their conclusions.

Jon Brock argues that it is important that psychology be considered a science, while neuroskeptic says that asking “Is psychology science?” is to ask the wrong question and the problem is that scientific knowledge is being conflated with true knowledge, an idea that can be traced back to Aristotle.

I feel that while we cannot say that psychology is not a science, we cannot say that it is a science either. But as far as I can see, nothing really hinges on how one answers that question except for bureaucratic purposes in universities and funding agencies who need to know which box to put it in.


  1. StevoR : Free West Papua, free Tibet, let the Chagossians return! says

    Easy test – what category does the local library file it under?

    Or, heck, these days what does Wikipedia say?

    Or does it follow the scientific method of hypothesis, experiments, and judging these by observed evidence?

  2. Enkidum says

    SteveR – all bad tests. The “scientific method” isn’t (there are all sorts of ways of doing science that aren’t hypothesis-testing – as I pointed out in the comments to Mano’s article last week, Darwin wasn’t hypothesis-testing on the Beagle, and Galileo wasn’t hypothesis-testing when he pointed his telescope at Jupiter), and the whole point is that the precise status of psychology is under dispute, so your local library and wikipedia’s opinion might change.

    At my university you can get a doctorate in psychology from either the arts or science faculties, with slightly different requirements in each case. I agree with Mano that ultimately it’s the wrong question, but practically speaking, it makes a massive difference for obtaining funding. (It might actually be an advantage – in Canada there are three main funding agencies for research, one for Arts & Social Sciences, one for Science & Engineering, and one for Health. I have had, directly or indirectly, funding from all three, which isn’t possible for most researchers in, say, sociology or physics.)

  3. consciousness razor says

    It is similar to art. I would argue that it is impossible to create demarcation criteria that determine what is art and what is not art, but we leave it up to the experts in the artistic community to make the final judgment, even if we may not agree with some of their conclusions.

    A necessary condition is that it’s created (not necessarily by a person) as opposed to a naturally-occurring object or event. That is simply what is meant by saying it’s artificial in construction, or there is an artificial context or meaning as in the case of “found art” or “musique concrète” or a bit of “performance art” that sneaks some drama into ordinary “real-world” events. This gets more confusing if a theist wants to claim everything in the universe is the artwork of a god, but they could still refer to what I’d call “natural” things as god-created or whatever, and the rest of us are at least partly responsible for being co-creators (or “procreators” even) of a distinct subset of everything there is, so in practice this doesn’t make a big difference (even though the language they’d use could differ dramatically).

    I’m not sure it’s impossible, but you’re at least right that whether there are other requirements is still an open question. Of course you can have aesthetic experiences about natural events (e.g., a sunset), so you might metaphorically call it “art;” but we can coherently distinguish between how we experience a thing (our response to it) and what kind of thing it is. There do not need to be subjective turtles all the way down. The question of whether it is or isn’t good art is also separate from whether it is or isn’t art, so if you’d argue that (the value of the art) is what’s impossible to determine, then it’s a lot harder to argue than you probably think. The “expert consensus” isn’t what I’d say determines that; but it’s a good guide, because these are the people who have thought about it the most based on a long tradition of other people thinking about it for a long time.


    For categories this broad and complicated, like science or art, I’d say a “family resemblance” kind of definition tends to be more useful (and people wouldn’t obsess over it so much) than drawing up a strict set of necessary and sufficient conditions. People might think falsification is all there is to it, for example, and then like you say it’s simplified so much that things like string theory or psychology can be tossed out prematurely. (Of course, a lot of psychology is falsifiable, but not all.)

    I tend to agree with Neuroskeptic. It’s an important point that the question is often not so much “is it science” as “is it physics,” so the answer doesn’t actually matter. It is more than just a bureaucratic issue, though, isn’t it? For example, take Richard Carrier’s efforts to make history more scientific by introducing Bayesianism. Whether or not it’s true in general that “history is a science,” to the extent it deals with empirical claims, there are better and worse methods of evaluating the claims, the better ones (with no exceptions I’m aware of) being precisely the “scientific” methods people are trying to define.

  4. Cuttlefish says

    Psychology, because of its unique history, is a tent big enough to contain both science and non-science (and a healthy dose of pseudoscience). Combining the questions of philosophy, the methodology of natural science, and the baggage of spiritualism, psychoanalysis, parapsychology, feel-good self-help horsefeathers, and worse, there are excellent contributions in both theory and application… which will receive little or no attention from outside the close circle interested in such things, while the latest from Dr. Phil defines psychology in the Oprah-viewing public’s eye.

    Relatedly, psychology is broad enough that experts in one area may be utterly ignorant of other areas; one media-savvy go-to scientific psychologist would not be nearly enough to make a dent in the public perception of psych as a playground of touchy-feely bullshit.

  5. trucreep says

    I’ve often thought that our understanding the brain and human behavior is in the “infant” stage at this point. We are on the verge of some serious breakthroughs I believe, but I think where we are right now may contribute to not wanting to commit it as a science.

    I believe as our understanding of the brain and behaviors sharpens, you’ll see the professions of psychology and psychiatry merge into one. For most cases, the two are absolutely needed, working in tandem with each other. I also believe mental health is tied in directly with overall health, and so it will be interesting to see how the ACA addresses that. I know they’re looking at having “behavioral screenings” at your PCP, and that very well may be a good start.

  6. Mano Singham says

    I agree that the ‘family resemblance’ idea is perhaps the most useful even if it cannot be precisely pinned down. It is how we make most judgments about categories.

  7. trucreep says

    I think a big factor in this too is that psychology is very personal at this stage. Meaning, the standards or mores are just starting to be established. So we all know having a healthy balance of HCL and LCL is important for every person, but how you define “healthy” is different for each person. I think Dr. Phil capitalizes on this, in that he knows the “standards” and doesn’t go much deeper than that.

  8. says

    … the problem is that scientific knowledge is being conflated with true knowledge…

    That’s kinda inevitable, when knowledge derived from scientific inquiry consistently turns out to be the most reliable knowledge we have.

    …we cannot say that [psychology] is a science either.

    Then what CAN we say it is? A pseudoscience? A religion? An art form? The field of psychology as a whole really doesn’t behave like any of those things — it behaves like a science, with observations, experiments, and at least honest attempts to offer testable hypotheses. Psychology is, at worst, a “soft” science, and that’s mostly because the object of its study, human minds, are so complex, idiosynchratic and unique that repeatable observations and experiments are much harder (if not impossible) to do. But none of that means we can’t accept it, and benefit from it, as a science when it’s practiced competently and honestly.

  9. trucreep says

    Agreed. I think because it’s still a relatively new field we’re only beginning to build a strong foundation of case studies and solid, replicated hypotheses. I think the majority will come to regard it as a science as time goes on, and more structure is built.

  10. mnb0 says

    “nothing really hinges on how one answers that question”
    I disagree. It’s true that nothing changes in psychology whether we call it science or not. But I maintain we can learn something – and that that something is often useful – from psychology, while we exactly learn zilch from even the most liberal form of creationism.
    “Psychology is not a science either hence ID should be taugh as well” is something I’d rather avoid.
    I’ll read your series, but for the moment it is enough for me that psychology looks for empirical data, compares them with theories and hypotheses and tries to formulate predictions that can be tested. That’s something the study of history also does btw. There is no single form of creationism, including theistic evolution, that does.
    But perhaps your series will make change my mind.

  11. mnb0 says

    Everything you write in your series The Logic of Science applies to economics, psychology and the study of history bar one: mathematics. But “psychology is not science because it can’t be expressed by means of mathematics” is rather a lame argument as mathematics is a language with it’s own grammar and all. One would have to make clear why the language of mathematics is such a good criterion. Moreover huge parts of the Evolution Theory would not be science. So imo the fact that psychology hardly makes use of math (sometimes statistics is involved though) makes clear this branch faces some problems physics manages to avoid, but is by far not enough reason to dismiss it as science.
    You confirm this in part 6 with

    “So some believers tend to try and devalue the insights science provides by elevating what we can call truth to only those statements that reach the level of mathematical proof,”

    even if the context is different.
    Also note that Freud and Jung aren’t very popular anymore among psychologists.

    “The reasoned consensus judgment of science”
    Psychologists have been doing a nice job in the respect last three centuries – far better than economists who have digged themselves in deep trenches. But it’s my opinion that scientists doing a bad job – they are to be found even among physicists; I point at Fred Hoyle rejecting the idea of the Big Bang – can’t be a reason to dismiss an entire field.
    So, in line with reader Jeff, I also have two questions. You write:

    “But it has proven problematic for the soft sciences, where there is no such unanimity.”
    1. can you prove that such unanimity is a priori impossible in psychology?
    2. to what extent the lack of this unanimity is caused by scientists doing a bad job? Here I am specifically thinking of economists, who don’t seem to do their best to formulate predictions that can decide or at least determine the limits of their competing pet theories, let alone to try to integrate them into one unifying concept.
    Psychologists at the other hand do, so my conclusion is that it’s a branch of science indeed.

  12. mnb0 says

    Finally let me present one hard fact we have learned from psychology. Complimenting kids – at least up to 20 – increases their self-convidence and is as such a more effective way to influence behaviour than threats and punishments. I know of a boy, back then 13 or 14, who used this fact to manipulate his younger sister to do a dull household job for him. That’s questionable from an ethical point, but his investigating mind should be praised.
    Apparently psychology is not even that soft anymore.

  13. Mano Singham says

    That fact may not be as hard as once thought. The recent work of Carol Dweck suggests that while process praise can help children towards greater achievement, trait praise has actually a negative effect.

  14. Mano Singham says

    Where did you find the phrase you quoted that “psychology is not science because it can’t be expressed by means of mathematics”?

Leave a Reply

Your email address will not be published. Required fields are marked *