Sunday, May 29, 2016

Falsifiable statements are not philosophical disagreements

I already mentioned this on Twitter:
But I thought I'd get a few more details in on the blog. I'm always a bit confused by arguments that are basically: We haven't figured it out, therefore we'll never figure it out. I've mentioned this elsewhere [1]. Just imagine someone arguing this about the nature of the Sun in the 1500s. Manu Saadia (whose new book Trekonomics is coming out in a couple days, and I plan to get it based on what I've seen him saying about it so far) got in an argument with Noah Smith about whether sociology should be scientific. Saadia fell into this line of reasoning:

Saadia says: "I don't believe social objects function and can be observed like physical objects."

This is a falsifiable hypothesis (the evidence would be the existence of successful theoretical and empirical sociological research) -- therefore it is not a "philosophical disagreement". Disagreeing about how the sun works is not a philosophical disagreement -- there exist various models that have varying degrees of empirical accuracy. One needs evidence that social objects have the properties Saadia endows them with by fiat. Saying it is philosophy seems like an argument that he doesn't think he needs evidence.

I call this the fallacy of arguing from failure of imagination; just because you can't think of a way social objects function or can be observed like physical objects doesn't mean it doesn't exist.

Now I guess it is fine to set Saadia's hypothesis as the null hypothesis, and you could say my argument is an attempt to capture the null hypothesis myself. But that's just a mathematical convention in statistical tests -- which hypothesis gets to be the null isn't specified by the mathematical theory.

H0 = Social science exists
H0 = Social science doesn't exist

If you switch back and forth between the two, you'd probably find there isn't enough evidence to reject either [2].



[1] I'm actually pretty proud of the whole series of posts that come up when you search for "complexity" on my blog.

[2] I'm being charitable. There is plenty of evidence that some aspects of social science are falsifiable, measurable and empirically testable. I'm just too lazy to look up some examples right now. However, the blog that you're reading right now is at least one existence proof.


  1. How would you define "philosophical disagreement"?

    1. The complement of scientific disagreement in the set of all rational disagreements.

    2. How do you define "philosophical"?

    3. The complement of scientific concepts in the set of all rational concepts.

    4. I'm glad I didn't ask you to define "dog" - I imagine you would answered the complement of the set of non-dog animals in the set of all animals.

      Very instructive.

    5. Nah, there are more than just two different subsets of animals. Indeed, what you've stated is a bit more tautological:

      D = not not D

      "dogs are not not dogs"

      This differs from what I said:

      P = R \ S

      "philosophy P is the relative complement of science S in rational thought R"

    6. I did not say:

      D = not not D

      That is a more general statement than the one I made.

    7. This is pretty typical of the way you do things - misrepresent what people say hoping no-one notices.

    8. "complement of the set of non-dog animals in the set of all animals"

      ND = D^C

      ND^C = (D^C)^C = D

      Set theory complement is formally equivalent to logical "not" operator.

    9. You said I said not Dog. Whereas, I actually said "non dog animals", that is the set of not dogs relative to the set of animals, not relative to the set of everything else (which is what you said I said).

      Completely different.

    10. Sorry, "everything else " should be "everything".

    11. If dogs are a proper subset of animals (and they are), then there is no formal difference between

      D = not not D


      D = (D^C)^C

      regardless of the set with which the complement is taken; it could be animals, it could be everything, or it could be things that start with the letter D in English.

      A \ (A \ D) = D if D ⊂ A
      U \ (U \ D) = D if D ⊂ U

      Both of these are different from what I said:

      P = R \ S

    12. Here we go again.

      I did not say D = not not D.

      ND = D^C is a nonsense statement

      D = (D^C)^C is a nonsense statement.

    13. Ah, I see.

      So ^C is a mathematical symbol for set theory complement. I'll use unicode if that helps.

      ND = Dᶜ
      D = NDᶜ = (Dᶜ)ᶜ

      I thought the symbols ND for non-dogs and D for dogs were relatively self explanatory.

      The last piece is an example of an involution, and is formally equivalent to the mathematical logic operation "not", or negation ...

      + D = -(-D)

      So you did; you said something formally equivalent to D = not not D.

    14. I took D^C to be D intersection C.

      I said the set of non-dog animals which is the set of non dogs relative to the set animals which is what you said and is not the same as not not Dog.

    15. All this nonsense is because you don't want to have a conversation about what you said above. Why don't you just say you don't want a conversation. That's fine with me.

    16. And I said that since dogs are a proper subset of animals (D ⊂ A), you are making a distinction without a difference that you remain adamant about for reasons that are mysterious to me.

      I'm not entirely sure what conversation you wanted to have.

      Would it have been better if I had titled this post:

      "Falsifiable statements are not unsolvable"


      The statement "X is not amenable to science" is a falsifiable statement. The falsifying evidence is showing a scientific success. For example, the statement "gravity is not amenable to science" is falsified by the successful prediction and empirical measurement of gravity waves. The statement "social systems are not amenable to science" may not have been falsified yet (there are some empirical successes, Okun's law comes to mind), but that doesn't mean it can't be falsified. Assuming it can't be falsified, and is therefore an unresolvable philosophical disagreement, is just a lack of the ability imagine it can be falsified. And someone else's lack of imagination and optimism is their bag, not mine. I can completely disregard it, and no one else should take it seriously either.

    17. It's exactly the same distinction you made in your statements.

    18. "The statement "social systems are not amenable to science" may not have been falsified yet..."

      That's the point isn't it. Effectively, we've been trying to falsify social science for 5,000 years, without much luck. Whereas, the last 300 years has seen the rapid advancement of the natural sciences. It seems to me one lends itself to falsification a lot more easily than the other.

    19. No, it's not. Including your distinction, you said:

      A \ (A \ D) = D

      dogs are the complement of the non-dog animals

      I said

      P = R \ S

      philosophy is the complement of scientific rational thought

      You'd have to define another subset of animals to be equivalent, but you don't, so what you say is tautological.

      You could have said warm-blooded animals are the animals that are not cold-blooded, which does break down the same way (and is formally equivalent):

      WB = A \ CB

      The problem is that you said "non-dog" which is a misrepresentation by you -- it calls what I said tautological, when it is not. I didn't say science is the complement of non-scientific rational thought.

  2. Jason,

    "the evidence would be the existence of successful theoretical and empirical sociological research"

    strikes me as almost by definition impossible.

    Wikipedia gives the following definition for sociology:

    "Sociology is the study of social behavior or society, including its origins, development, organization, networks, and institutions"

    I don't know about you, but "origins, development, organization, networks, and institutions" seem inextricably linked to reasons and/or explanations to me.

    Tell me, how do you propose we empirically measure the underlying reason behind, e.g., the social development of money? How do we observe the reason for something in the first place? You can observe an empirical cause (e.g., heat causes ice to melt), but what about a social one? Can empirical evidence determine the reason for the existence of, for instance, a specific political institution?

    1. I'd say that empirically successful economics is successful empirical sociological research; economics is a subset of sociology (studying the exchange aspects of human behaviors).

      You ask "measure the underlying reason behind, e.g., the social development of money?"

      What is the underlying reason behind quantum field theory? How do you measure it?

      And there are lots of political institutions that derive from biological human behaviors -- our violent streak as human animals has been institutionalized as the military. Our social hierarchies, as government. I don't know a lot about those particular fields, but I imagine there are lots human evolutionary biologists and evolutionary psychologists out there who can connect the institutions we have today with aspects of human behavior.

    2. "What is the underlying reason behind quantum field theory? How do you measure it?"

      That's exactly my point. This question is 1) inherently non-empirical and 2) part of sociology and not science.

    3. My point was that the reason behind the existence of quantum field theory is at worst philosophy (assuming it doesn't follow from string theory [it does follow, but string theory may not be a correct theory itself], in which case it is simply a derivation).

      When you ask:

      "how do you propose we empirically measure the underlying reason behind, e.g., the social development of money?"

      I actually take a stab at it here:

      It predicted the existence of metal cowrie shells in China ... at least before I knew about them! That is an empirical success :)

  3. Jason, O/T: what do you think of this piece by Sumner:
    This statement in particular caught my eye:

    "If back in 1990, we wanted to know whether there were Higgs bosons or gravity waves, the optimal guess would not have been derived by asking a single physicist, but rather setting up a prediction market."

    1. Also, who's the thug in the scary looking mugshot (thug-shot?) you're using as your new avatar? =)

    2. I read that, and was thinking of writing a post on it because I think it illustrates exactly what is wrong with prediction markets (or market indicators of expectations in general). They're basically asynchronous polls of conventional wisdom. So if people have the wrong model (e.g. exchange rates or interest rates), the answer from a prediction market will be wrong.

      By the 1990s, gravity waves had been observed as energy leaking out of neutron star binaries, e.g.

      You don't have to directly observe something for it to exist. Atoms weren't directly observed until the 1980s, but they "existed" empirically since 1905 when Einstein came up with a way to measure Avogadro's number.

      Regarding the Higgs, the Higgs mechanism was fairly well established (it predicted the W and Z bosons along with their masses), but there are theoretical problems with a massive scalar particle like the Higgs (such as, Why is its mass not the Planck mass or the GUT scale?) -- implying the existence of physics beyond the standard model.

      So prediction markets somehow know about the physics beyond the standard model that allows a massive scalar to exist at energies below the unification scales?It's like saying prediction markets could have predicted quantum mechanics.

      (And that is me outside my friend's hangar. I got tired of the other pictures and thought I could use something that pops more, colorwise.)

    3. Thanks Jason.

      I knew it was you... it's a good photo, but it is slightly mugshot-like. I like it!