Still Searching For That Left Handed Gene

This is a short follow up to the post I had on the paper written by Lawrence Mayer and Paul McHugh. Something smells fishy about it. Two issues stand out to me: 1. It presented itself under the neutral auspices of “science,” as opposed to some kind of underlying moral theory which the authors won’t admit to; and 2. they pretend to “care” about the well being of LGBT people, and indeed, don’t say anything overtly nasty, but give ammo to sources who are more than willing to do the dirty work.

So we have this article entitled “Born That Way? A False Hypothesis” (from a project at Grove City College doing what I think to be good work which I enjoy) which seeks to use that ammo. In a raw philosophical scientific sense, the title of this article is fallacious. Yes, it may be true that we don’t have smoking gun evidence sexual orientation is fixed at birth (which I would admit, I don’t know or believe to be true). But still, absence of evidence is not evidence of absence.

I think the real “false hypothesis” is that proving “born that way” one way or the other is dispositive in an “ought” sense. Analogies can be tricky things because any time we make them we have to compare an A to a B or an apple to orange. But in terms of the etiology of the condition (not necessarily the propriety of the chosen acts), the closest analogy to sexual orientation is to that of handedness.

And by the way, we are still searching for the “left handed” gene. We thought we found it; but that turned out to be wrong.

And also, interestingly, left handedness and minority sexual orientations tend to correlate with both great talent as well as some negative things.

Please do be so kind as to share this post.
TwitterFacebookRedditEmailPrintFriendlyMore options

11 thoughts on “Still Searching For That Left Handed Gene

  1. If the central dogma is true, then of course everything is genetic. Experience has taught us so far, however, that, just as with everything else in life, there is a seldom a single, predictable cause for a given outcome.


    • Genetics is….complicated. While I have no idea what biology classes teach now (and given how many school districts have issues just teaching evolution, I’m terrified to look), but back in the day it was “Here’s Mendel, single-site D/r gene examples. You’ll do some fun squares and figure out percentages, and although we’ll say “Most things are far more complicated” THIS is what you’ll remember and think of when ‘genetics’ comes up”.

      Most things are…not single site. (Those are pretty easy to find anyways, so they’re low hanging fruit). Most things are not simple on/off — they have a chance of expressing, which depends on all sorts of factors (including environment, womb or otherwise) and sometimes just random chance.

      So on the one hand, there’s often things you can show are genetic just by doing things like twin studies, interviewing families, and doing a bit of statistics to show “This is unlikely to be by chance”. But the actual mechanisms? Sometimes you have to push past correlation (sometimes traits are strongly correlated with other traits, and you can zoom in on the wrong one), you have to deal with the fact that a lot of things are multiple-site with varying degrees of penetrance…..

      OTOH, with the human genome project and computers getting faster and faster (and storage getting cheaper and cheaper), well — we can finally start digging in a bit deeper and probing actual mechanisms.

      (Although if you want fun, try to imagine what a database containing a genome would look like — how HUGE it would be. Not imagine the fun of running searches, comparisons, and the like through a database that large. The computer scientists get a real workout, is what I’m saying.)


      • (I know I’m late to the party here.)

        First, it help to understand mixed strategies. In a complex world, you don’t optimize for fixed behaviors. Instead, you optimize for good probabilistic distributions.

        Second, it helps to understand that genetics is a (sort of) computational algorithm. But more, it is the seed that expands into a computational algorithm, itself based on computational algorithms. Except it’s not even that. There is no “meta.” There are no “layers.” It’s all mixed together, like a Lisp program where everything is a macro that takes macros and produces macros that take and produce more macros, loops of macros all the way down.

        Each turtle rests on countless turtles above. It is a cyclic digraph, more than sparse but not quite dense.

        But even that analogy falls short. After all, software systems are way more precise and predictable than the physical world. Whereas computer software tends to run in a very predictable environment, living beings exist in an utterly unpredictable world, an ever shifting fitness landscape, full of arms-races and non-convexity and non-smoothness and randomness and every possible thing that makes the math of optimization really hard.

        So the relationship between genetics and human brains and neuro-plasticity and “social construction” and so on, when applied to sex/gender — good grief, we don’t know anything. Let us stop pretending we know things.

        I know about my own body. I know how I feel. I know what gender transition did for me. This I know.

        If you wanna know more, you have to listen to people like me. After all, science is great, but the science leads us to nearly boundless complexity.


        • “If you wanna know more, you have to listen to people like me. After all, science is great, but the science leads us to nearly boundless complexity.”

          Which is why people that want to reduce everything to capital letter first principles (like Truth) bug the hell out of me


        • “Whereas computer software tends to run in a very predictable environment, living beings exist in an utterly unpredictable world”

          … I think that the internet is pretty unpredictable. But I will admit that software does sometimes have trouble understanding meatspace — “this is just a plan, it didn’t happen yet — no, you can’t name yourself Lord of an American City — people still live there, for now”


          • — By “computational algorithm”, I mean complexity in terms of Turing computation, not in terms of the internet, which is a gigantic mixture of asynchronous software systems and the humans using them.


              • So far no one has shown that QC can compute anything that a Turing machine cannot, only that a few very specific NP-intermediate problems can be solved in polynomial time.

                Don’t trust any positive claims about QC that do not come with specific algorithms along with formal proof. For example, the whole “adiabetic” thing [1] depends on two things: we can convert a general instance of 3-SAT into a very specific Hamiltonian form in P (unproven) and also that the gap between the two lowest eigenstates is sufficiently large to guarantee the computation won’t “jump states” during the “cooling” period (also unproven).

                (Unless of course someone has recently proven these things. I don’t keep up with the latest.)

                So anyway, [citation needed] all over the place.

                Read Scott Aaronson’s blog. (Which actually, I need to catch up on his blog.)

                [1] I know one of the “big names” in the adiabetic space. Actually, I think he has a crush on me, which is cute.


Comments are closed.