September 1, 2010

What are social science's accomplishments?

Jim Manzi writes in his article What social science does -and doesn’t- know that 'the social sciences have not demonstrated the capacity to produce a substantial body of useful, nonobvious, and reliable predictive rules about what they study—that is, human social behavior, including the impact of proposed government programs'.

In the article he says that social sciences have been relatively late in embracing controlled experimentation which is an essential method to settle debates about what works and what not. According to Manzi, even now that the experimental method is becoming more and more popular we cannot expect fast breakthroughs in our scientific knowledge about 'the human condition because of the high 'causal density', the number and complexity of potential causes outcomes of interest, in this domain. He concludes: "At the moment, it is certain that we do not have anything remotely approaching a scientific understanding of human society. And the methods of experimental social science are not close to providing one within the foreseeable future. Science may someday allow us to predict human behavior comprehensively and reliably. Until then, we need to keep stumbling forward with trial-and-error learning as best we can."

Is this a fair judgment of social science's accomplishments? Surely, social scientists have done many experiments and acquired much knowledge. On this blog I have frequently mentioned examples of social scientific knowledge on topics like self-determination theory, growth versus fixed mindsets, development of expert performance and outstanding achievements, stereotype threat, priming, and so forth. But I have to admit, these topics, although very useful,  are examples of relatively fragmentary knowledge about human functioning. A 'scientific understanding of human society', of course, lies on a much higher aggregation level. But I would say that social science can be quite useful even if it can not yet 'predict human behavior comprehensively and reliably'.  I agree, we are very far from understanding the whole of human function and human society. But I wonder if this is a fair criterion for determining the usefulness of social science. Often, you don't have to understand the whole in order to be effective. To name just one example, we know a lot about the do's and don'ts of effective teaching.

Still, I agree with Jim Manzi that in many of the complexer tasks and domains of life we need to stumble forward with trial and error learning.

What are your thoughts?

12 comments:

  1. It's a good question! A lot of social science is about understanding, see http://en.wikipedia.org/wiki/Hermeneutics

    As such, much of it only shows results if you study social sciences yourself: you read what others wrote, and understand more.
    That understanding doesn't need to come in the form of facts. In that way, you could argue that social scientists understand their area: the problems can't be reduced to simple formulas. Or if they can, then they are boring, not interesting.

    Let's see what the next decades bring.

    Stephan

    ReplyDelete
  2. Self-determination theory, growth versus fixed mindsets, development of expert performance and outstanding achievements, stereotype threat, priming, etc... these are great finds but are on an individual level.

    Society is more about interaction between individuals and, in my humble opinion, this is virgin territory from a scientific point of view.

    I have a passion for Socionics which is a personality type theory focused specifically on interaction between types. Based on Jung's type theory as Myers-Briggs' theory, it started from the social interaction part and then moved to explain the types.

    This is virtually unknown in the West, the first respectable book appeared only this year.

    In my Internet meanderings I have yet to find anything remotely close to Socionics.

    I'm not saying that Socionics is The Best, but it is a start. There are some attempts to formalize it in order to better study it from a scientific point of view, there are also some mathematical demonstrations for the 16 types by one socionist with mathematical background but it is very difficult to study society... the complexity is quickly escalating.

    A quick example, if you have a dual couple you can have a phenomena called dualization that is very similar to the Flow state BUT if you ad another personality to this couple, the dualization might never appear because the third personality will probably disturb the flow of information HOWEVER, if the personality of the third is from the same quadra you again have a very interesting type of collaboration within the group formed.

    ReplyDelete
  3. Hi Peter, I agree these examples are all on an individual level. But examnples on a higher level do exist (here is an example: http://ow.ly/2yonX).

    I am a bit skeptical about the approach you mentioned, but who knows... perhaps it will develop to something very valuable.

    I am rather hopeful about approaches which involve network science, complexity theory, social dilemma research, etc.

    ReplyDelete
  4. Why are you skeptical about Socionics? What triggered you skepticism?

    ReplyDelete
  5. fair enough! I've seen enough to know that personality typing is very thin ice. :)

    ReplyDelete
  6. by the way, that Kahn academy you mentioned some time ago ... is awesome!

    ReplyDelete
  7. Khan Academy is indeed awesome in what it is but it is way beyond awesome in what it could become.

    Have you see the Vision and Social Return video?

    ReplyDelete
  8. No, haven't seen that one. I'm beginning with some math and chemistry

    ReplyDelete
  9. Oh... you're going to love it! :)

    My mind was blown at the part where he talked about monitoring video watching in order to find places where the video can be improved.

    ReplyDelete
  10. I've watched it now. I agree, it's very powerful

    ReplyDelete

Enter your email address:

Delivered by FeedBurner