[Most Recent Entries]
Below are the 20 most recent journal entries recorded in
[ << Previous 20 ]
[ << Previous 20 ]
|Friday, May 11th, 2012|
|Digital brain confusion
I wanted to share this
because it's so full of very characteristic confusion.
Adherents of the computational theory of mind often claim that the only alternative theories of mind would necessarily involve a supernatural or dualistic component. This is ironic, because fundamentally, this theory is dualistic. It implies that your mind is something fundamentally different from your brain – it’s just software that can, in theory, run on any substrate.
This is true but a bit confused, and I cite it for context. For now, all I'll say is that a a convection current is a convection current regardless of the fluid instantiating the current, so there's a sort of dualism there, too, but only that between logical existence (i.e. for any whole number and there "exists" a square of than number) and instantiated existence (street lights "exist").
By contrast, a truly non-dualistic theory of mind has to state what is clearly obvious: your mind and your brain are identical.
One could say the photograph I took and the patches of positive and negative charge on my hard drive are identical, and there would be a sense in which this was true, but not a very interesting sense. If we were to take this seriously as the true ontology of the photograph, then the sequence on my camera's SSD card and that on my hard drive are not two instances of the same photograph, but completely separate things. Again, there's a sense in which this is true, but it's not the sense that matters
to us, and for good reason. Likewise, there's a sense in which the mind and brain are identical, but it's not a sense that matters to us. In fact, if we refuse to recognize the senses in which the mind and brain are *not* identical, we are also refusing to recognize the senses in which minds *matter*.
As an example, there's no obvious moral relevance to moving around atoms in a brain, and indeed it happens constantly. But if they move in such a way that the mind is extinguished or impaired, then it is a tragedy and/or a crime. Clearly, we care about what a brain *does,* not the specific matter of which it's made.
Now, this doesn’t necessarily mean that an artificial human brain is impossible – it’s just that programming such a thing would be much more akin to embedded systems programming rather than computer programming. Moreover, it means that the hardware matters a lot – because the hardware would have to essentially mirror the hardware of the brain. This enormously complicates the task of trying to build an artificial brain, given that we don’t even know how the 300 neuron roundworm brain works, much less the 300 billion neuron human brain.
This has a thread of truth to it, but ironically, the way its wrong is embedded systems programming is hardly closer to artificial brain design than computer programming, because it still implies creation of a mind-containing brain by a programmer. It's worth noting that the genes for creating the brain are, for all their complexity, orders of magnitude less complex than the machine that they set up, and which continues to create itself for years. If genes don't have to program the brain, then neither do we. If we did, then I would be the first to agree that I can't see how we would ever create hardware brains.
For one thing, the brain itself isn’t structured like a Turing machine. It’s a parallel processing network of neural nodes – but not just any network. It’s a plastic neural network that can in some ways be actively changed through influences by will or environment.
Exactly. We set up the plasticity, and the mind creates itself.
What follows is the author throwing the word "emergent" around without defining it (a problem because there's several ways of using it, and it's a like playing whack-a-mole to try and counter the confusions they introduce). Fortunately, I think I can ignore those paragraphs as irrelevant and skip to the conclusion.
It’s possible to build computers that can learn and solve complex problems. But it’s much less clear that there’s an easy road to a computer that’s geared towards the type of emergent properties that distinguish the human brain. Even if such properties did emerge, I’m willing to bet that the end result of a non-human, sapient intelligence would be very alien to our understanding, possibly to the point of non-comprehension. Electric circuits simply function differently then electrochemical ones, and so its likely that any sapient properties would emerge quite differently.
Again, this is true if we make the same mistake as the author of assuming that artificial brain creation is meaningfully like computer programming. However, if the designers set up an artificial brain like our genes do, then it's not nearly so hard. Still incredibly challenging, yes, but something I expect to happen in my lifetime, assuming we continue to work at it. Basically, we have to learn enough about the gross-level organization of the brain and the behavior of its constituent neurons to generate something like the information topology and dispositional state of a fetal brain, and then the brain does the rest. Complex as they are, genes aren't magic, and they just aren't sophisticated enough to create something conscious. Neither can we. But we can create something that becomes conscious like any other mind, and that mind would be very, very like our own.
So I would reverse his conclusion almost exactly. Something very much like us is almost the only thing we could
create, at anything like our current level of understanding.
|Monday, January 30th, 2012|
Here's the latest rankings of the various presidential administrations' GDP growth numbers. I rank by real per-capita GDP annual growth percentages rather than real gross GDP, because higher population growth also increases GDP, even if society isn't really getting any richer. I count from Q2 of the first presidential year to the final quarter in which they served. Thus, Clinton's second term is considered to be Q2 1997 to Q1 2001.
2.40 per capita
2.38 per capita
1.77 per capita
1.44 per capita
1.38 per capita
1.28 per capita
1.02 per capita
0.82 per capita
Obama I (so far):
0.76 per capita
G. W. Bush I:
0.67 per capita
-0.23 per capita
G. H. W. Bush I:
-0.63 per capita
G. W. Bush II:
-1.57 per capita
|Friday, October 8th, 2010|
|Probably unnecessary to say
Gold is in a bubble. It is not a safe harbor asset, it is an asset that, if purchased right now, will have large negative ROI in the medium term. It will lose half its value. Those advising investors who need to avoid risk that they should backstop themselves with gold are doing a terrible thing.
|Wednesday, August 4th, 2010|
|A good day
Though it looks like a stay will probably be issued
. As ever, I'm conflicted about this (legal/constitutional) method of overturning gay marriage bans, but I would feel most positive if the stay was lifted for at least a while so people can get married. Besides the humanitarian end of allowing people to pursue happiness, it gives other people more chance to see that - contra the innuendo of anti-gay bigots - the only real consequences of allowing gay marriage are all positive. Then when time comes again to vote again, voters will correct their mistakes.
|Friday, July 30th, 2010|
|On the occasion of revised BEA numbers
Per-capita GDP growth for each presidential term of the last half century.
Because the first quarter of a new presidency is divided between incoming and outgoing, I also include a parenthetical value that shows results of allocating the first quarter to the outgoing president.
Reagan I 1981-1984:
Reagan II 1985-1988:
Bush Sr 1989-1992:
Bush Jr I 2001-2004:
Bush Jr II 2005-2008:
Obama 2009-present: [extrapolated to a full term]
0.5 (1.9) [1.2 (6.3)]
I'll come back a little later to do a bit of political analysis. Of course, I'm a skeptical regarding the impact of presidents on economic performance, but it's clarifying to understand what results would
suggest if one insisted on a strong causal relationship.
|Monday, June 21st, 2010|
|Not the best debate
I was sort of glad that Brand didn't waste rhetorical time on cost comparison
because it's easy to merely sound like someone who doesn't want to pay for the necessary alternatives, when trying to convince environmentalists who aren't already well disposed toward nuclear. Instead he talked about arithmatic and "always on". This is good, but you have to back it up. Jacobson shows the nice graph in which average solar, wind and hydro can cover everything, which very much made it look as if those could actually be
"always on". Brand starts to respond by saying it's a sunny day and a windy night, but since he doesn't follow it up, it makes it sound like he thinks the numbers are best-case. Of course they aren't - Jacobson wouldn't do that and anyway I've seen them before. The problem, though, isn't that they're exaggerated or best case, it's that they're not worst case. That
was what Brand meant when he was talking about the two cold weeks. There would be many times when renewables would supply much more than needed, and other times when they would supply enough. But then there would also be times when they wouldn't supply nearly enough for extended periods of time - weeks or whole even seasons, not hours or days - and that would be crippling.
I also think it might have been worthwhile for Brand to clarify his bit about storange to make it clear that most plants have enough onsite storage for high grade waste to last well beyond their lifetime: there's not that much cause to worry about fissile materials being trucked all over.
|Saturday, June 5th, 2010|
|Some truths about transportation subsidies.
To bang on the same drum as I have in the past, I figured I'd cite some current figures.
Federal income from the gas tax in 2008: 29.6 billion
Federal land transport expenditures in 2010: 72.7 billion
State income from gas taxes in 2010 is no more than about 60 billion
State land transport expenditures in 2010: ~110 billion
Local land transport expenditures will be another ~130 billion.
Totaling up, that means that gas taxes cover up to 90 billion of the >310 billion our governments spend on automotive transport, or no more than 29%. Average farebox recovery for transit systems in the US was about 31% in 2007 and has increased since then, meaning that mass transit is actually less subsidized than automotive transit.
Some specifics: Both the BART and CalTrain have higher than average farebox recovery (BART by a lot, CalTrain by a little bit). The LA MTA is a smidge below average at 30%, but still better than driving. DART in Dallas is the worst at 12.7%, no doubt because Dallas is not at all laid out for transit. Hong Kong's transit agency is self-sufficient at 100%.
The idea that transit is an inefficient government boondoggle while the automotive industry is a free market marvel is total BS.
|Sunday, April 11th, 2010|
|"The Civil War wasn't about slavery"
It just occurred to me that denying that the American Civil War was (primarily) about slavery is morally equivalent to Holocaust denial. I'm not talking about problematizing narratives in a way that recasts the issue in other terms (e.g. 'the Southern aristocracy and economy was intertwined with slavery in ways that meant that the loss of slavery would bring down the rest') without trying to deny the centrality of 'the peculiar institution' to the conflict. That's necessary to more fully understand the ways in which various threads of human motivation interact.
But the South seceded to defend slavery
. There was little or no mention of economics, honor and other such lacunae cited by deniers in the declarations of causes of succession issued by the state conventions that decided on secession, and other similar documents from appointed representatives and etc. Compare with the huge
preponderance of arguments related to slavery. Almost the only things in any of them are defenses of the legality of secession and arguments that secession is necessary to protect slavery.
Considering the number dead in the conflict, and the enormous - almost unimaginable - injustice of antebellum slavery, I can only compare the compulsion to disregard the words of the prime movers of secession with the motivations of those who wish to pretend that the evidence of the Holocaust is exaggerated and/or manufactured. It's repulsive, and should not be socially tolerated.
|Saturday, April 3rd, 2010|
|Sustainable Energy - Without all the hot air
Really quite good
I just read pretty much the whole associated book in one sitting (again derailing a writing day :-( ). It's an analysis an engineer can love. There's very little economics in it, but its really great to see what the real possibilities are, abstracted from quibbling over cost projections. The short answer is that the 80% solutions boil down to nuclear and possibly solar farms in deserts closer to the equator. Other possibilities can help, but those are the 20% solutions.
Of particular note to me was the quantity of recoverable fissile material plausibly (though perhaps not economically) available, which is many times larger than I'd appreciated. Both electric cars and rail can get more savings than I'd appreciated*. In one rather pleasing section, MacKay confirms my most optimistic estimate of savings from living vertically: climate control energy losses in "terraced" (e.g. how-house) homes without any insulation are about equivalent to the most efficient detached homes. A modern high rise home would require no more than about half as much energy to heat/cool than an equivalently sized single family home. Another item that impressed me was how horrible currently extant "clean" technologies are. My favorite was Earthrace's "eco-boat" that runs on biodiesel, which was about as energy efficient in transporting passengers over distance as the Queen Elizabeth II cruise ship, as long as the QE2 was towing several more empty ships of its size behind it.
I remember at the end of the 90s I was disgusted with the environmentalist movement for being idiotic about figuring what would actually be good ways of tackling the various problems. Well, it would appear that it hasn't come as far as I'd thought. Sure, there are increasing numbers of converts from Greenpeace and whatnot, but while the back-to-nature hippies have definitely lost control of the movement, the gadgeteers that have supplanted them aren't helping that much either. Also, all the "every little bit counts" stuff turns out (much as one might expect) to achieve just what it says: a very little bit. If everyone driving switched to hybrid cars, if everyone unplugged their phone chargers, etc etc. it would all amount to single-digit savings as a percentage of overall energy use. The big savings:
#1 By a large margin is dense living, which enables the following large-impact items;
1.1:Freight and transport efficiency
1.3:Farmland efficiency (because it doesn't push agriculture out of the most fertile land)
#2 Less meat-eating - which would be both healthy and economical to boot
#3 Electrification of passenger transportation
#4 More efficient home appliances
#5 Reduction in air transport - I hesitate to put this here, because I think a proper carbon price would allow people to decide how "worth it" air travel is, and it's not really a guarantee that air transport would really drop enough for it to be considered such a large savings.
#6 Reduction of packaging
All of that probably still only adds up to 30-40% savings, and I think it's fair to suppose that we're going to continue to find more things to do with electricity and whatnot, so I don't think we can expect to get much below 100kWh/day/person in the mid-to-long term. In the mid-term, we'll supply 70% of that with either coal or nuclear fission. We'd better get on the latter now, because it's got waaaay fewer issues than coal.
*Electric cars seem to be able to save about 80% of transport energy relative to gasoline ICE, comparing favorably to the negative 150% 'savings' of some hydrogen powered prototypes. Electric rail saves up to 98%.
|Sunday, March 28th, 2010|
There are other gasoline use externalities (smog, geopolitical harms), but the use fee I propose below exerts plenty of incentive force by significantly undercharging efficient cars for road usage and overcharging the inefficient, so there's no need to get into all the difficulties of estimating the values of those.
Also, the pigovian tax is likely to drive up the overall cost of gasoline and other fossil fuels more than the $0.04 cent/gal tax to end users would suggest, because emissions in the refinery and distribution process would drive up costs to producers. I don't really know how significant this would be, but those who protest the "exorbitant" costs imposed might consider that if the overall carbon-related costs are (let's say) $0.10, that means that we're already using the equivalent of 2.5 gallons of gasoline just to get one gallon of gasoline into the gas tank.
One final item is that my proposals don't address one very important aspect of environmental degradation that is currently handles really poorly: water use. For that, I think the government should auction various kinds of long-term environmental extraction leases to the various resources and establish arbitration courts in which bid-winners can sue one-another for environmental impact losses. if an oil platform leak hurts a fishery, the fishery can recover their costs. If farmers want to divert so much water they'll destroy the fish stock, they'll have to contemplate the costs of reimbursing the entire cost of lost fishing income every year subsequently. I have a sneaking suspicion they may decide it's a bit cheaper to just use drip irrigation. Again, we can refund half the income to the public to whom the natural world belongs, and set aside the other half for mitigation efforts.
|If I had my druthers
We should have a pigovian tax of $40/ton of CO2, with half of the revenue refunded and half used to impact mitigation efforts for AGW or whatever. Then there would initially be an actual-cost fee added to gasoline to pay the real lifecycle costs of roads - for Texas it would be something like $2/gallon, but it might be cheaper or more expensive in other states depending on weather, utilization density and vehicle fuel efficiency. The pigovian emissions tax on a gallon of gasoline would be fairly small: <0.4$/gallon. The usage tax would be so much higher it would have a much greater impact on efficiency. There's a disconnect, however, because that efficiency doesn't mean people are using the roads less, just using less fuel while using the roads the same amount. Raising rates to spread the costs over smaller and smaller amounts of fossil fuel wouldn't work for very long, besides being unfair. Eventually, we would need to either work out a toll system, or make a mileage fee part of registering a car. The latter would be technologically simpler but socially much more difficult, since it would mean people would have to come up with significant lump sum* amounts every year. Mass transit systems would need to drop their subsidies as well, of course, which would double fares on many systems, without the below measure.
Meanwhile, there are always going to be some perversity to real estate markets that works against the best land uses because the purchase prices will always be determined by potential rather than actual worth, but ongoing costs like property taxes can at least favor efficiency. Property taxes should be based on something like economic value per hectare rather than economic value per hectare plus improvements. Currently the market sets values somewhat speculatively, meaning the most potentially useful plots are extremely expensive, dissuading those who might have wanted to improve them from doing so, and essentially rewarding those who put land to poor use. Property taxes depend on market prices, so they reinforce the same trend. I think econometric analysis can separate improvement values from land values fairly accurately, and even getting close would be better than what we have now. The portion of value deriving from proximity to mass transit could be applied to the costs of running the mass transit, reducing fare increases. Essentially I would regard these as something like land use fees rather than property taxes in the general sense.
I think these four things would accomplish the vast majority of what needs to be done to drastically improve the environment with the least cost and impact on individual liberty.
*My thumbnail estimate is that it would be about ten cents per kilometer, which isn't bad if you only drive 3000km a year, but a fellow with a fairly moderate 20 minute commute of 20km or so would drive about 10,000 km/y just for work: a $1000 lump sum payment just to get to and from work.
|Sunday, December 13th, 2009|
|Quick outline of the compatibilist/incompatibilist arguments.
Does free will exist? Is my perception of me
making my choices an illusion?
To these questions the compatibilist says "yes" and "of course not". Embedded in the common construction of the free will/puppet dichotomy is the idea that in order for a choice to be free it must be uncaused. The compatibilist replies "no, in order for a choice to be free it must be caused by salient features of one's self
." If a choice is made at random or through coercion, trickery, etc, then of course it isn't free, or at least its freedom is attenuated. The incompatibilist will usually reply that if the contents of self are determined by the outside world, then they are not really ours. There are a number or responses to that, but the general thread is "our selves are our
selves, however they eventuate." There are other incompatibilist complaints, but those usually amount to insisting that the traditional incompatibilist definitions of things are pretheoretical base intuitions and are thus axiomatic.
|Saturday, August 29th, 2009|
|For my birthday
Kat gave me Richard D James' Ventolin single, a house, and SNICKERDOODLES! They are yum.
|Tuesday, June 9th, 2009|
Most people reading this probably know that I have every expectation that we will build true human-type artificial intelligences in the next 2-4 decades. Those who have suffered through my longer monologues on the matter know that I worry about what are we to do with the inevitable mistakes we'll make. Evolution produces many, many creatures that live short, horrible lives for every one with favorable mutations, and I expect that humans attempting the same engineering task will do the same. How are we to treat entities - people, arguably - with all manner of developmental disabilities, emotional problems and so on? Will we regard them as experimental failures and terminate them? If we decide they have full rights a biological human would have in the same circumstances (as would be my inclination) then who is responsible for their care throughout their arbitrarily-long lifespans?
Another related issue: If rogue (or not so rogue) states start making intellectually-crippled robotic warriors and workers, are we morally obliged to preserve them? Will there be some cutoff point below which we'll consider entities to be machines rather than disabled people?
We don't have to have answers to these questions yet, of course, but it will help to start trying to engage our intuitions on the matter beforehand, when there aren't any high-stakes controversies encouraging tendentious conclusions. For example, if the US military is the source of intellectually-crippled robot soldiers, then we Americans might be strongly inclined to ease the sense of cognitive dissonance by setting the bar for humanity just high enough so that we don't have to condemn our own government. Further, if we establish some sort of consensus beforehand, then governments have some guidelines of which to avail themselves when deciding what kinds of projects to pursue. Already democratic governments perform polls on the ethics of robotic weapons systems, so if the people being polled had definite opinions on the matter then we can avoid moral horrors in much of the world.
|Friday, June 5th, 2009|
As Chris Bradford notes
, the Texas Department of Transportation recently did a study to determine which roads pay for themselves in gas taxes. Not a single road or highway met that bar; a few paid close to half and most paid far less. The shortfall, of course, comes from discretionary funds.
We don't generally pay the real cost of mass transit, but neither do we pay the real cost of driving. My daily rail commute would cost approximately $14.20 unsubsidized. If I drove, however, I would use about 3 gallons of gas, or about $14.50 unsubsidized (i.e. with gas tax rates sufficient to pay for the roadway). Add to this the insurance costs of adding those miles to my car as well as the increased likelihood of personal injury/death, and rail starts to look a lot cheaper. If one adds the savings from not having to have an additional car (let's say $12 for an inexpensive late model car), then rail starts looking like it might be a lot cheaper after all.
And that's before we even start talking about network effects or land scarcity.
|Thursday, May 21st, 2009|
|Tuesday, May 12th, 2009|
|Identity in an age of cognitive enhancement
We have been accustomed to treating humans as a special ontological class - or perhaps the archetype of the class - because of differences in the manners in which we have identities. Two folding chairs don't have the same identities, usually, but neither do we care very much. If one made a chair that constantly cycled its constituent atoms, we'd have no trouble transitioning to treating its identity like that of a river, which we reify in terms of its persistence rather than substance. If two such chairs plashed together and reformed, many observers wouldn't mind saying that there's no fact of the matter as to which chair became which.
Humans, however, have always had a very discrete, non-fungible character that has formed the building block of moral reasoning. That we do not exchange properties overmuch and retain our narrative persistence has been critical to grounding our basic concepts of identity and cause attribution. In coming decades, however, we will find more and more ways to augment the minds that evolution gave us, until we of the future will be vastly different than we are today. If I and my brother, for example, both get the same implant that dramatically increases our intuitive grasp of, say, econometric calculations, have we become in some way the same person? What if we had a great number of identical enhancements at the same time, covering everything from dance coordination to art visualization? Unlike the chair, it can be quite unsettling to have to find a place to stand once we have to contemplate persons in a manner more like rivers.
But we always have been rivers, and only now is the looming future forcing us to see that fact, like studying geology made us realize rivers are only so permanent and identifiable. Do we fight to hold back those realizations, or do we face the fact that we are all of us changing all the time, subject to the contouring pressures of chemistry or accident or even consistent self-discipline? She who can tread those waters need never worry about finding a place to stand again.
|Tuesday, April 7th, 2009|
|Thank You, Vermont.
Now they can't say it hasn't been passed democratically. Now for California to get on to repealing Prop 8. And the feds to repealing DOMA. And DADT, for that matter.
|Friday, March 13th, 2009|
|Another note on water usage
I recently ran into a post on a rural agricultural newsletter website regarding the need to educate 'city folk' on the need for water conservation.
Per capita average indoor water use - including leaks - in the US is about 70 gallons in fairly high-range estimates
. Another 100 gallons goes to watering the lawn, washing the car and so on. Per-capita total water usage is about 1,350 gallons/day. The unaccounted for 1,180 gallons per day are industrial, commercial and, by a large majority, agricultural.
This map of water usage per capita by county
excludes industrial and agricultural usage*. The City and County of San Francisco, which has negligible industry or agriculture but plenty of commercial, uses about 102 gallons/day per capita, which includes indoor, outdoor, commercial, etc. If S.F. cut its water usage by 55 gallons per person (i.e. more than half), it would save approximately 135 acre-feet per day or about 50,000 acre-feet annually. This is no more than a sixth of one percent of California's agricultural use.
Meanwhile, if agribusiness found a way to save 1% of the water that it currently uses in irrigation, it would save enough water for another 2.5-3 million City folk.
*Note that California's total
per-capita water usage is more than 1,000 gallons/day, so if they didn't exclude industrial and agricultural usage, some of those counties would show thousands of gallons per head.
|Thursday, March 12th, 2009|
|Hormones and babies
In order to evaluate the common claim that women of a certain age experience a hormonally-driven desire for babies, it is necessary to 1) unpack the possible meanings of the assertion and 2) decide what the truth of those meanings would imply.
Rarely does the casual observation come with a great deal of explicit detail, the communicator relying instead on common experience to both substantiate and elaborate. In an audience listening to the inevitable stand-up comic’s reference to such conventional wisdom, there are certainly as many different understanding of the joke’s content as there are different people listening, but to two friends of long standing the gap between intent and reception is presumably much smaller. Rather than giving up interpretation for lost, however, one can stereotype certain interpretations by tracing the route from frequently-perceived patterns, through different understandings how hormones and people work, and to the different parsings of the assertion if question.
The classic hostile (either to the claim, or to female agency, depending on one’s perspective) reading is that desire for babies is elemental to being a women on account of female biology, full stop. This has the appeal of defining gender in an immutable way that obviates the need to address complexities in womens’ relationship to childrearing, thereby stabilizing culture and simplifying a number of social questions*. Probably no one very thoughtful holds this view in its simple form, but it remains a perhaps unconscious contributor to other interpretations.
A more common source of the observation is that as women age, they become more likely to want children, and that this desire becomes more pressing and visceral at times. This is fairly uncontroversial as a statistical claim**, but the attribution of these trends to hormones is more dubious. It does help explain why there is so frequently a feeling of personal compulsion attendant on the desire*** for babies, and it’s well known that hormone levels change with age, leading to a hormone-as-scapegoat interpretation. At the same time, the attribution is probably somewhat of a screen: by attributing the desire to something external to the self – hormones – one is saved from saying that women of a certain age just are some certain way. At the end of the day, why exactly women tend to start desiring babies at a certain age isn’t the important part of the observation; the interesting part is the remarkable consistency with which age changes womens’ manner of viewing children. Call this the hormone-as-rubric version.
At the opposite end of the interpretive spectrum from the hostile version is when a speaker is obliquely referring to individual womens’ increasing awareness of the time limit for child bearing and even child rearing. This need not be explicit; even fairly young women must include the fact of hormonal time limits when they begin to make long term plans. In this version, the hormones have very little influence except through gender-neutral practical considerations.
What if hormones really were a primary cause, as in the hormone-as-scapegoat hypothesis? Well, hormones are just chemicals; rather blunt instruments that can’t drive a mind into a certain state. Could they excite a part of the brain associated with nurturing impulses? Likely, they could, but the object of those impulses would necessarily be a product of the individual woman’s mind. It seems difficult to imagine a hormone driving its host to nurture a baby but not, say, a pet or a potted plant. To create greater specificity, another hormone might create a coincident escalation in libido, connecting a greater desire for intercourse with a desire to nurture. This seems plausible, but once again it’s easy to see how the individual woman could respond to both urges without human infants coming into play. The missing connective tissue, of course, is culture. It is culturally “normal” for womens’ desire for babies to increase with age, increasing the likelihood that any given woman will interpret the several changes in her drives as being a growing desire for babies.
This blends into hormone-as-rubric explanations in which the individual’s life situation becomes more amenable to having children as she ages. Women in their thirties are generally more established financially and socially, have the confidence of greater knowledge of the world, and know that their cutoff date for bearing children is approaching. Further, their friends and family of the same age are having their own children, creating a sense of community in childrearing. And, of course, there may be explicit social pressure to become a mother.
Obviously, individuals all have different experiences with this time. Some mothers are no better off financially, or cannot expect any child rearing aid, or don’t have friends who are having children, and so on. However, most women experience do experience many, and few could fail to note the regularity with which women of a certain age evolve a desire to have kids. Hormones could certainly figure in this intersection of causes, amplifying the effect just a bit more, and then when people refer to hormones making women want to have babies, they’re speaking a variety of truth. The hormones are neither necessary nor sufficient, but they at least have a statistical part to play and so are a useful stand-in for the whole ball of wax.
Which leaves us with the totally practical version, in which the referent is entirely the practical effects of hormonal changes that end reproductive years. This would apply to men as well, of course, if they’re committed to a single woman. To the extent that this is true, society does at least give a nod to men of a certain age evolving a desire for children. However, it is also culturally normal for men to be rather footloose, and anyway men-with-hormones is not a standard trope. Men are, according to the standard construal of the gender, unaffected by any hormonal drive except lust, which would only drive the standard man to seek a younger wife when the current one approaches menopause. In fact, any instinct toward nurturing is so strongly associated with femininity that a man living to standard gender norms would likely regard making concrete a desire for childrearing somewhat emasculating. It’s most often framed as a concession or a mishap at the hands of a (scheming) woman.
One could hypothesize that the apparent increase in female desire for children is a product of practical changes in a woman’s apprehension of her ability to undertake motherhood combined with the suppressive effects of male gender roles on the concomitant increase in men’s ability to undertake fatherhood. One need not avail oneself of any appeal to hormones or even female gender roles**** at all. That doesn’t mean there’s no intuitive reason to suspect a role for hormones , just that there isn’t a prima-facie case for the conventional wisdom that hormones make women want babies, except in the most prosaic, uninteresting sense.
Given that a primary cause of the old saw’s popularity is subscription to - and approval of - unhelpful prescriptions about gender roles, it would be best to retire it. Women who want and are ready to have children will keep becoming mothers without being told it’s normal for their age and gender, and some of the men worried about their masculinity might even come to covet the natural respect due a good parent.
*One does not, after all, waste a lot of time considering if being held to the ground all the time is constricting because gravity just is, whether we want it or not. If biology renders certain facts of life similarly irresistible then we can stop worrying about whether they’re good or not.
**Though it would be interesting to see what the actual numbers are
***”Desire” here meaning only a contemplation of the attractions of something, not an incipient intent to acquire it.
****Properly regarded, the female/male gender role binary isn’t separable, so I’m engaging in a bit of rhetorical excess here. That said, the binary only need exist in the minds of men to cause the cited effects, so it’s true enough.