SMBC: Social Science

Occasionally Saturday Morning Breakfast Cereal captures an important insight, in this case, people’s attitudes toward the social sciences.

Click through for the red button caption:

My attitude toward the social sciences is that they are quite capable of being scientific.  They’re not always, but then even the “hard” sciences have their lapses.

On the one hand, what social scientists are studying exists at a higher abstraction layer from the more natural sciences.  It reminds me of this old xkcd:

Six people are shown representing six scientific fields. ... Psychologist: Sociology is just applied Psychology. Biologist: Psychology is just applied Biology, down to Physics which says good to be on top. Math is shown far to the left not even noticing the others.
Click through for source:

Each layer adds additional complexity, and uncertainty.  Physicists can often achieve six sigma certainty in their results.  Biologists can’t.  And it becomes hopeless in psychological or sociological studies.  The social sciences currently attempt to achieve p-values of .05 or lower, which means a 95% chance that the result is what they think it is, far lower than what can be achieved in the natural sciences, but a simple fact about the practical epistemic limitations they face.  (Crowd sourcing services like Mechanical Turk can help, but not to bring the results to the levels enjoyed by physics.)

But the biggest issue is that, unlike the natural sciences, social science isn’t really studying timeless issues, but human behavior, and human behavior can change, particularly after humans have heard the results of social or psychological studies and taken that information into account.  To a large degree, this makes studies of human behavior a moving target.

Alex Rosenberg, philosopher and self-labeled nihilist, after considering this information, concluded that the social sciences are hopeless.  He thinks they’re at best entertainment.  (A category he also relegates the humanities into, including presumably his own field of philosophy.)

Does that mean the endeavor is hopeless?  I don’t think so.  I think the issue comes from a binary view of knowledge: either we know something or we don’t.  But that’s an unproductive view of knowledge.  It leads many to conclude that true knowledge is impossible.

A more productive view is that knowledge is any reduction in uncertainty.  With that conception of knowledge, observations that reduce our uncertainty about things, from complete unknown to broad probabilities, is better than no knowledge at all.  Of course, less uncertainty is better than higher uncertainty, but any reduction of uncertainty, that is, any increase in certitude, is worth the effort.

This all seems like fairly common sense, so why do people resist it?  I think the last caption above captures it.  The social sciences are studies about us, and the results often clash with people’s deeply felt intuitions, intuitions that are behind many traditional views of humanity.  The issues above on lack of certainty and timelessness simply provide an excuse for people to ignore the data and assert their own intuitions.  When we get into fields like economics, political motivations, conscious or unconscious, factor heavily into it.

My favorite field, neuroscience, sits on the boundary between biology and psychology.  I suppose we shouldn’t be surprised that its results are often tainted in the same way that psychological studies are.  Studies about how our mind works are the ones most intimately about us.

28 thoughts on “SMBC: Social Science

  1. Nice post Mike. I think what is really ironic is the fallacy that somehow specificity and truth are synonymous. For an instrumentalist, specificity provides a higher level of certainty. All too often, that higher degree of certainty is a compromise which has “absolutely” nothing to do with the true nature of reality.


    Liked by 1 person

    1. Thanks Lee,

      For me, it comes down to how well our models predict conscious experience. If they do, then we say they’re “true”, at least to some degree. Are they really true, where by “true” we mean corresponding with ultimate reality? I don’t know that that’s something we can ever answer. We never get ultimate knowledge of reality. We only ever get our own conscious experience and our predictions about it.

      If those predictions are testable, at least in principle, we call them hypotheses and the models that produce them theories. If not, then we’re in metaphysics land. Of course, today’s metaphysics may be tomorrow’s science.


      1. Even though as a discipline metaphysics may possess a greater explanatory power over physics, it lacks specificity; and it is the specificity which provides the certainty that the human psyche demands, regardless of whether that certainty is predicated upon a fairy tale or the true nature of reality.

        I keep referencing general relativity and the fabric of spacetime as the quintessential example. Even though GR has great predictive prowess, at a fundamental level there are so many things wrong with the theory that I find it difficult to believe that Einstein himself would still support it. The theory is really cool; it’s an ingenious intellectual construction, one that deserves a noble prize. But does it reflect the true nature of reality? Not a chance.



        1. What leads you to conclude that GR doesn’t reflect the true nature of reality? I think most physicists do think either it or quantum theory will eventually need to be updated, although given that both theories have been tested to an astounding degree, the changes to the mathematics will have to be pretty subtle. Of course, that doesn’t mean a dramatic shift in narrative might not happen.


          1. Here’s the answer to your question Mike: GR is a synthetic a priori judgement, a judgment that expresses a condition on the possibility of reality, not the reality itself. Synthetic a priori judgments are the bedrock upon which all of our cherished theories are built and synthetic a priori judgments are the only means we have at our disposal to construct those theories. That is a fulcrum concept and that concept is a well kept secret, a secret that nobody within the scientific community wants to discuss because it challenges the authority of the priesthood.



          2. Lee,
            I can see the logic for saying that a scientific theory is synthetic a priori knowledge. The goal of a scientific theory is to be an a priori understanding of its subject matter, which can then be used to deduce (predict) experiences. But construction of such a theory might happen through a priori reasoning, but also typically involves a posteriori activity, not to mention a scientific theory has to be testable, which seems like an a posteriori exercise.

            That said, it’s totally possible I hopelessly misused the philosophical terms above. My overall point is that science is relentlessly pragmatic. It typically uses whatever type of reasoning works, in whatever combination of analytic, synthetic, a priori, and a posteriori that produces results.


          3. No, you expressed the philosophical terms well enough Mike. But don’t forget, so-called “results” are always subjective. My thrust is that there is no way to demonstrate this so-called fabric of spacetime through any form of a posteriori analysis. The entire notion of GR is predicated upon synthetic a priori judgements. Sure, the mathematical models we constructed work very well, but the accuracy of the predictions are not 100 percent. And for any theory to reflect the true nature of reality there cannot be any exceptions.

            What science does with GR is no different than what Bernardo Kastrup does with his intellectual construction of a cosmic mind he calls mind at large or M@L. The only difference is a matter of scope. Both M@L and GR are grounded in synthetic a priori judgements, judgements that express conditions on the possibility of reality, not the reality itself. I put no more credence in GR than I do Bernardo’s M@L for the same fundamental reasons, neither model will stand up under the scrutiny of analysis.



          4. One final anecdote on GR Mike: What is all too often overlooked is that in layman’s terms, the notion of space being like a fabric is an analogy. Once that is understood, one can frame the question of whether GR relativity is true or not in the correct context. All analogies are true for the “thing”, which is the analogy; but they are not true for the “thing-in-itself”, which is the true nature of reality. Hope that helps to clear things up because when it comes to expressing a condition on the possibility of reality using analogies, all analogies break down; end of story.



          5. Thanks Lee. I see the point you’re making now. It isn’t the mathematics you’re attacking, but the metaphors and analogies used to describe the theory. (Max Tegmark once remarked that physicists refer to these narratives as “baggage”.) I actually agree that those metaphors might eventually turn out to be wrong. Nothing that I’m aware of that depends on them being more than metaphor, such as wormholes and the like, has so far been empirically validated.


  2. I don’t consider it a 5% error rate in experimentation, or false intuitions, which mandates that our mental and behavioral sciences remain soft. It’s more Mike’s proposal that these fields concern us . How might scientists manage to objectively explore the nature of something this personal? Obviously physicists don’t have to deal with such an issue. Thus the fundamental science of psychology, which should naturally feed them all, has not yet been able to develop any broad generally accepted theory regarding our nature. Here I speak of theory which would remain effective beyond standard behavioral changes, and even various species of creature. Without such theory, modern psychologists seem to remain entrenched in micro theories which are unable to do much for fields such as sociology, as well as neuroscience’s modern venture into consciousness.

    So how might psychologists power through this problem to develop effective general theory regarding our nature? As I see it, psychology’s main jailer (and thus the rest of them), exists as the social tool of morality. These shackles should need lifting in order for psychologists to finally be free to develop theory which is amoral (as is standard for hard sciences) which would effectively reduce our nature to an idea which is experimentally successful. (In my opinion such theory should suggest that we’re all self interested products of our circumstances, and regardless of our incentives to deny this.)

    I believe that the field of psychology (and thus the rest) will ultimately be freed by a field that’s in even worse apparent shape, or philosophy. If / when a reasonable group of philosophers is able to develop various generally accepted principles of metaphysics, epistemology, and axiology which scientists find helpful for their own work, then this band should grow. Here we’d have a field which is more on par with science that could then leave traditional philosophy to thus provide something beyond culture or art. Furthermore these principles should provide soft scientists with a means from which to develop effective general theory regarding our nature, and so help them all.

    I did a post and was interviewed on the subject about a month ago:

    Liked by 1 person

    1. Eric,
      Physics does have its theories that disturb our sense of self and place in the universe. The classic example, of course, was Copernicus’ heliocentric theory, which moved us from the center of a universe that literally revolved around us, to an insignificant mote of leftover stardust in a vast void. Subsequent revelations have only hammered on that ancient conceit.

      More recently, the many-worlds interpretation of quantum mechanics threatens our very sense of a unique self, which I suspect is a reason so many people detest it. And as I noted in the post, there’s a lot of resistance to neuroscience for similar reasons. So, there’s a long history in science of trampling on our intuitive conceptions of ourselves.

      I need to swing back to that interview. Unfortunately, long videos are tough for me because of the time commitment. You guys didn’t happen to do a podcast version did you? Listening is easier since I can do it while walking.

      Liked by 1 person

      1. Mike,
        Copernicus’ heliocentric theory is a perfect example of a person who took down a vast paradigm in order for our “hard sciences” to rise. I’m proposing no less than this for modern soft sciences. Then you bring this full circle by mention non testable bullshit like the “many worlds” interpretation of quantum mechanics. So apparently even the field of physics now fails given the failure of philosophers to develop generally accepted principles of metaphysics, epistemology, and axiology from which to filter out such nonsense.

        Yes Liam and I did actually do this interview as a podcast as well. You can access it by putting on your earphones, sticking your phone in your pocket, and taking your walk! 😜


        1. Copernicus’ theory was non-testable for several decades. The only thing it had going for it throughout the 16th century was that it was more elegant than the Ptolemaic model. During that time, many people reacted to it similar to how you react to MWI.

          Youtube videos stop playing when the phone screen cuts off (I don’t have a Premium subscription), and even if I hold it, I haven’t found videos to be reliable once I’m outside of my home’s WiFi bubble. Hopefully I’ll be able to watch it soon.

          Liked by 1 person

        2. Hey Eric,
          I watched the interview. Again, as I noted in my email the other day, you present much better than I would in that situation. I’m pretty familiar with your ideas, so there wasn’t much surprising for me.

          I do think your categorizing economics as a hard science is something a lot of people will disagree with. It seems to suffer from all the ills I listed in the post, and there are always people pointing out its many failures to predict recessions, inflation, and other economic maladies. (I think those criticisms are unfair, but then I think a lot of the criticisms of social science in general are unfair.)

          Liam pressed you on quantum physics and causality. Your bottom line is that it’s deterministic. You pretty vehemently dismissed one of the deterministic interpretations above. Do you feel any better about others, such as pilot-wave, or Hossenfelder’s embrace of superdeterminism? It seems like every interpretation comes with a cost, in the case of deterministic ones, proliferating worlds, non-locality, conditions in the big bang leading to synchronized results in entangled particles, etc. Just curious if you’ve given this any thought.

          Liked by 1 person

          1. Mike,
            Well I didn’t do as well as I’d hoped, though the video certainly could have instead been quite embarrassing. Fortunately Liam was great. Real time interviewing (and certainly on camera) is a learned skill that I’m not yet versed in. This was my first such lesson, though I would like more.

            I didn’t quite say that economics was a hard science, but rather that it was on the harder side of things. And why would I assert this? Because the field has broad generally accepted theory which seems to work. You might have noticed economists over at Rationally Speaking explain how the true work that economists do isn’t all that controversial. Apparently it’s mainly just that media fueled accounts of the field get drenched in politics.

            As far as not being able to predict what will happen with inflation very well and so on, even presuming “perfect” models, there should generally still be too many unknown variables to plug in. At least note that economists make far more effective predictions regarding the micro issues of individual businesses and people. This is probably because better information generally exists for this sort of thing.

            Economics may be contrasted with our most central behavioral science, or psychology. From the days of Freud to various modern proposals, this field hasn’t yet developed any broad generally accepted theory regarding our nature. The point of my post and interview, is that I consider the social tool of morality to prevent the central behavioral science of psychology from accepting that happiness constitutes the value of existing, and even though the side behavioral science of economics has essentially been permitted to take this step. Thus I believe that we’ll need a respected group of philosophers to agree upon something like my single principle of axiology in order to help psychologists raise their science to the level that economists have already achieved for their field. This still unaccepted theory is sometimes referred to as “psychological egoism”.

            I didn’t quite state that quantum physics functions deterministically. I said that either it does function deterministically in the end, or instead it reflects a void in causality. I consider the second option to be an extremely effective definition for “magic”. It’s my own metaphysics of naturalism / causality which binds me to the idea that even though we don’t understand what’s happening, causal dynamics should exist to potentially understand anyway. Thus I consider Heisenberg’s wonderful uncertainty principle in an epistemological rather than ontological capacity. (In fact, I accept no theory ontologically.)

            I don’t understand pilot wave theories, or certainly Hossenfelder’s superdeterminism (which I believe she’s joked about being one of two people in the world who does). And it could be that I don’t understand the many worlds interpretation either. I simply read a little about these ideas on the blogs sometimes. But isn’t it true that you’ve read Sean Carroll’s latest book, and listen to his podcast from time to time? Perhaps you could let me know if I’ve got MWI wrong?

            As I understand it, matter isn’t thought to function as a wave or particle, but effectively both. We see this in that the more precisely we measure its state at a given moment in one regard, the more that our measurements will be confounded with variability in the other regard. Thus we find a distribution of probabilities rather than any exactness of existence as things get more precise.

            To me that’s all fine. I don’t think we should expect to measure a particle perfectly if it’s not a particle, or a wave perfectly if it’s not a wave. So I assess the whole situation here under the heading of “ignorance”, and conveniently presume that causality applies anyway given that this would foster dynamics which could conceptually be understood. Otherwise I’d classify myself as a supernaturalist in this regard.

            As I understand it, “many worlds” proposes that whatever happens in our world is mandated to happen exactly as such, and so is deterministic, though each of the unfulfilled probability potentials also occur by spawning off into separate universes which are exactly like ours except for the noted difference. So as I understand it, within a cubic millimeter of space there should tend to be gazillions of new universes spawned each second given the unfulfilled probabilities associated with general material dynamics within. So entire universes are spawned instantly out of “nothing”, and this conveniently helps us interpret the variability of our measurements to be deterministic? That’s why I’ve assessed this proposal as “über magic to the nth degree”.

            Let me know wherever you perceive me to be mistaken, of course.

            Liked by 1 person

  3. Speaking on behalf of SMBC, what do you mean “occasionally”? 😀

    That’s one of my favorite xkcd cartoons. (And exactly why I love mathematics so much. Its purity.)

    As for the soft sciences, it’s mostly a matter of hugely complex systems and their variable space exploding. And, as you said, in many cases, these systems are evolving over time — the laws of physics and mathematics famously don’t do that. There is also that, even with weather, for instance, gathering sufficient fine-grained data — even if you had a cogent theory for it — is a big problem. (And when you data has to come from people’s mouths, fuggedaboudit.)

    Liked by 1 person

    1. Yeah, when your research data comes from asking people questions, it inevitably taints everything with a cloud of subjectivity. To some degree, it can be controlled and compensated for with various statistical techniques, but it’s never as “clean” as measuring phenomena with an instrument. (I quoted “clean” because I’m sure a lot of people working with such instruments would point out that the data is often not that clean at all.)


      1. It’s quite possible that such large systems cannot be solved in any effective way. There being too many variables is bad enough, but the subjectivity and graining of the data make it even worse.

        Hari Seldon was wrong. “Psychohistory” isn’t possible.


        1. Strangely enough, there’s a movement called cliodynamics, which resembles psychohistory in some ways, although it has far less ambitious goals.

          It’s interesting that even in the stories, Seldon needed a backup plan in case it all fell apart. (Which of course, it did.)

          Liked by 1 person

  4. I would add one thing to that comic: This random, out of context sociology paper agrees with my own personal intuitions. Therefore, my personal intuition is now a proven scientific fact.

    Liked by 1 person

  5. David Sloan Wilson is currently doing a series of conversations on his podcast about integrating evolutionary thinking into “Economics, Public Policy, and the Third Way.” It’s excellent. I just finished transcribing a few notes from his conversation with the economists David Colander which had something germane to your post. Here they are:

    –> “It’s so natural when you are thinking of the economy to think of complex evolving systems. In fact, economics was very much at the foundation of thinking about evolution. For example, look at Malthus operating there and seeing examples that Darwin used. So that kind of thinking was part of the overall theme. The question then was, what are you going to do with evolutionary thinking? In biology, they developed a way to talk about particular things. Economics moved in a different way trying to have an algebraic theory that would allow us to specify things and put them in an equation. Economics moved to an attempt to mathematise something, whereas biology just talked about things and expanded from there. Then what happened was that it became very hard to fit things into the mathematics that was available, so economics lost its complexity and the richness of evolution as part of the discussion and focused instead on what we can hold constant in service of these other equations. Walrasian general equilibrium theory was very much a part of that. [Walras, a talented mathematician, believed he proved that any individual market was necessarily in equilibrium if all other markets were also in equilibrium. This became known as Walras’s Law.] Walrus looked at when equilibrium will be arrived at, whereas complexity looked at it and said we don’t care if equilibrium is to be arrived at because it never will be. There will always be change in the system, and the question is how will change be adapted to, and how will people react to that change, and what brings about positive and negative change. That’s exactly the questions being asked by you [David Sloan Wilson] now, so that is economics going back to its roots rather than being something new.”

    –> “What has made this new discussion of economics possible is algorithmic science as opposed to equational science. By that, we think in terms of processes, and solutions to particular problems, and algorithms are the basis of that. Algorithms tell you how to solve a particular problem rather than provide an overall theory. There’s a big change going on within economics that I think is moving it towards a complexity science.”

    –> “Economics went in a physics direction, not in a biology direction. In order to become mathematical, it was really emulating Newton’s laws of motion, in which individuals are akin to atoms. However, even Newtonian physics had to yield to more complex theories. In just the move from a two-body problem to a three-body problem, our mathematics doesn’t work and you have to move to an algorithmic method of step-by-step estimations and recalculations.

    –> “At the end of the day, we’re going to have something like weather prediction where we have models, lots of them, and data, lots of it. The models are continually tested against the data and refined. Big data and machine learning are playing a big part in this highly iterative process. Evolution hit this barrier very early with Mendelian genetics going from 2-locus theory to 3-locus theory. [I.e. genes in 2 or 3 locations act on one trait.] So, mathematics is both highly essential, and highly limiting, and for that reason it needs to be humble in its aspirations. When you model something, you know that its a cartoon of a very small piece of the real world and unless you are calibrating it again and again and again, and integrating with other models that shine their light on another small piece of the world, you are misusing the role of mathematics in complex science. In evolution, we were never misled to think that there could be some grand system of equations which we would call a theory.

    Liked by 1 person

    1. Thanks Ed. That podcast sounds interesting.

      I’ve often thought the weather forecasting model was a good one for social sciences to aim for. It admits there are inherent limits in our ability to predict systems, and focuses on reducing uncertainty rather than permanently banning it, which isn’t in the cards.

      In truth, it’s not even in the cards for the harder sciences. Even physics measurements come with margins of error in their measurements. It’s just that they’re more able to reduce things to simple variables than sciences like biology and higher.


Your thoughts?

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.