Zero is a special number, but one is too
So she does exist after all.
It is often said to people that they have “unrealistically high standards” for the person they would marry, the idea being that looking for such a person would doom them to a life of foolish loneliness. And without evidence, that could be true. But what if that changes?
The hunt for smart women
I have always said that I want to marry a woman who is mentally sharp. People ask me what I mean by that and how I can gauge this, and although my going definition is quite intuition-reliant, it still works quite well: somebody that has a “spark of reason” to them, where you can “see the cogs turning” immediately upon being given information, but rather than her thinking momentarily to have a quick low-resolution response ready, what you are seeing is the process of thinking through all possible mental models in real time. I also hypothesise (because I am this way) that somebody with this spark is also somebody with a rich inner world; somebody whose brain comes up with realisations about complex domains of life (like the philosophy behind good relationships) in the shower, and perhaps also notes those down to attempt to map out the complexity of things.
Here is the problem: the amount of women of this calibre that I have encountered in my life is probably countable on one hand (without reusing fingers). All of them I met in or before high school, and as these things go, all of them have dropped out of my life with me never having made moves at them. Given that I’ve interacted with many, many more people and that I’ve long lost contact with those then-girls-now-women (and since this was high school and before, I can’t even really remember if they qualified), I know – for all intents and purposes – zero women of interest.
Now, although women’s IQ curve1 has a smaller variance than men’s – meaning it is harder to find very dumb and very smart women than it is to find very dumb and very smart men – it is still nonzero at the high end. The problem with a population distribution, however, is that it cannot give any guarantees about a finite sample’s distribution. In this case, that finite sample without guarantees consists of all the women I know and meet. In fact, the lack of guarantees gets worse: IQ being perfectly normally distributed is only a approximative model, which is most obvious at the tails where supposedly there is still a nonzero probability of having 1000 IQ, so at some point it breaks down at the population level. Secondly and more importantly, the people we meet aren’t actually sampled from the population distribution, nor are they independently sampled: who we meet depends heavily on who we know and where we go.
The above analysis holds for any property we we desire in a mate. Hence, although it might exist in the population, it’s not obvious that it exists in the pool an individual samples from. That’s where desparation sets in: am I, unbeknownst to myself, sampling from a pool that doesn’t have the quality I desire in it? Or am I sampling correctly and is it just rare? This nuance is lost when people say you should “lower your standards”: is it because there are no people in the world that meet the standards, or is it because you’re not looking in the places where people do meet them?
Zero is a special number…
Brett Weinstein, professor of evolutionary biology, came up with the dictum that “zero is a special number”. In his case, this refers to evolutionary standstills: cases where a species lacks a certain quality which, once a single individual evolves it, will immediately start dominating the population and bring the species unbeforeseen prosperity. For example, the first human civilisation to develop a combustion engine could economically and militarily outcompete any other civilisation that hadn’t. Yet, as long as nobody had invented it, everyone was on the same playing field in that respect. Zero is a special number.
In my case, with respect to sampling from an unknown distribution, zero is also a special number: as long as a value has occurred zero times, it’s impossible to discern whether it cannot occur or whether it can but has not yet. This holds at two levels: at the level of the population model (we don’t know where the model overestimates probability) and at the sample level (we don’t know what is present in the non-uniformly-chosen subpopulation we sample from).
But just like the evolutionary case, one is equally special and groundbreaking: it only takes one occurrence for the entire picture to change. It tells us simultaneously that this value does actually exist in the population and th not more than onceat we are actually sampling from a viable subpopulation. In the case of people, we get even more information: where we encounter them gives a clue about similar future encounters, and – again, because we don’t sample our acquaintances independently from each other – we can probe them about where to find more like them.
I just had this zero-to-one event happen to me.
…but one is too.
I was at my first academic conference (I won’t namedrop in this article, but it doesn’t take long to figure out which one), meeting droves of people who are brainier than anyone I normally encounter on a daily basis. As is now customary in academia, the majority of the researchers I met were women.2 Apart from coming to the presentations of their work, I also shared meals with them and went sightseeing, and there was never a dull moment in there.
All of the exchanges I had were valuable in some kind of way, both “on the pitch” conversing about our work and “off the pitch” hanging out and talking about aspects of life like compatibility in love. I met more interesting people during that week of conference than I met in five years of university study. I’m very happy that my self-confidence has grown sufficiently at this point that I can successfully approach and talk to new people, which is far removed from where I once used to be.3
This essay came about due to one conversation “on the pitch”. If you were an outside observer, you would have thunk standing far away that these were just two people gesticulating in funny ways, and from up close that this was nothing but a highly technical conversation about abstract spaces. You would have never guessed that below the surface, it had a profound personal impact on me.
This was the first time in recent memory that I had found a woman with the spark I described at the start of this article. A mind visibly wrapping itself around a highly abstract mathematical idea and then playing with it by going back and forth with interesting propositions.4 Some might think this is a big nothing burger: obviously smart women exist. Yet, this is a specific kind of smarts and eagerness to engage with ideas, and I can discern that it is there because it is exactly the trait I always missed the most in women I’ve dated in the past. Coincidentally, I found her beautiful as well.
In defence of being picky
I talked about this exact subject with one of the other women I went sightseeing with. Unlike most people to whom I mention that I have certain intelligence standards, she didn’t judge my needs as unrealistic. She did rightfully point out an interesting problem that I will have to contend with on this journey: as an engineer in computer science (and specifically NLP), I have only a narrow domain of subjects based on which I can fully deduce the “compute” – to use a term from our field – available to someone. The critical reader will read this as me actually being a moron in most subject domains, which would indeed make my entire search hypocritical because there is no use having a wife who is smart in a domain we can’t together be smart in. I will make three counter-remarks to this.
Firstly, just because my detector for someone’s sharpness of mind is domain-specific, doesn’t mean my or the other’s intelligence are as restricted as the detector. The point of marrying someone with intelligence is precisely that it is transferrable, particularly to the realm of the relationship itself. If one person has to dumb themselves down too much when negotiating anything, any conversation will be like pulling teeth, taking two steps back for every step forward.
Secondly, the view that one can be very intelligent in a given domain and a dimwit in all others confuses knowledge with reasoning ability. Going back to transferrability, one of the hallmarks of having a high capacity for reason is that “software updates” – understanding novel things – happen quickly. In other words: even if the person is not knowledgeable in a particular field, it takes little explanation to get them to a good baseline of mental models (and, as a side note, you would be surprised at the amount of people who lack the curiosity to even want to explore new ideas, even if they could).
Thirdly, I should point out that I actually do have a broad field of interest and understanding: for example, my high-school degree was in classics with a supplement of math and science. Additionally, in Belgium, an academic bachelor’s degree in engineering science (the first higher degree I obtained) provides students with a broad, generalist education in applied technical sciences, even if its purpose is to prepare students for an academic master’s programme that is very specialised. Many people, including some of my fellow graduates, find it strange that I, a computer scientist, was mandated to take courses from electrical, civil, chemical, mechanical and materials engineering. They only think of immediate applicability as a justification for studying many complex subjects, failing to realise that it is the complexity of the subject that is the point, not the subject. You’re only peripherally learning the subject, whilst actually training the creation of mental models.
I have retained a good chunk of knowledge from my generalist educations, and I enjoy foraging into the more advanced corners of biology and physics, even though neither is related to my area of expertise.
In sum
I don’t believe that I can only detect if a woman is right for me if she is specifically a researcher in NLP. I’m not sure of this though, but I would hope that I wouldn’t overlook her when she comes along.
But at least I now know that the probability of that happening is non-zero, if I look in the right places.
Using the concept of IQ might come off as shallow, as if I am fixated on a single number (like others would be on e.g. the size of the body or its parts, quantified in your metric of choice) rather than the person. The point of referring to IQ is simply that it provides a readymade mathematical model to speak on intelligence as a stochastic variable. I’m not fixated on the particular number; it’s a quantitative proxy for the qualities I have described. ↩︎
Fortunately, I matured out of my “shy hopeless romantic” era half a decade ago, so I have no issue treating them as if they were just an extension of my male homies without involving any romantic interest. (It might seem strange that it could be any other way, but it’s not so obvious that a man and a woman who share their curiosity and interests won’t develop some kind of “more than friends” dynamic when forced to meet in a box, far from home. It definitely takes non-zero maturity to keep your professionalism, as is attested to by the many official and unofficial romances that spawn in workplaces all over the world.) ↩︎
I attribute it in part to a reduced impostor syndrome (I have an undeniable track record showing that I am actually competent at thinking), in part to a reduced self-consciousness about my physique (I thank my consistent workouts for that), and in part due to an increased apathy to the results of any interaction (I have nothing to lose and something to gain, let’s try it). ↩︎
Once again, lest the malicious reader think this is a misogynistic piece about the mental faculties of the sexes: my surprise was not “Wow, you’re a woman and you can do this, there don’t exist many of those.” but rather “Wow, I have found one of the women that can do this, it doesn’t happen often that I cross paths with that group.” I don’t believe that the sex of an individual is related to their ability to reason. The point being made here is that I have found difficulty sampling from a group I am very optimistic exists just as much as the equivalent male group. ↩︎
