The returning soldier effect

27 Jan

A friend told me a joke the other day. A squaddie awaiting his train home after Armistice 1918 was interviewed on the platform by a radio reporter.

And what’s the first thing you’ll do when you get home?

Take the missus upstairs for a good seein’ to!

Hmm. This is for a family audience. What’s the second thing you’ll do?

Take this flamin’ pack off me back and get out of these ruddy boots!

An earlier conversation had touched on a phenomenon we’d independently heard presented in New Age terms, invoking an ‘intelligent cosmos’. That’s the Returning Soldier Effect: a spike in male births after major wars.

Since then, others had told me that returning soldiers and their women are indeed, as the joke suggests, keen to make up for lost intimacy. Doubtless, I’d said. So? So, replied these friends – social scientists some of them – the more we have sex in any monthly cycle, the more likely it is we’ll conceive a boy.

Not the ‘intelligent cosmos’ then. But is it true that more nookie means more boys? That part is not easily verified or refuted. Two other statistical phenomena come together in this account:

Survivors – at least among British WW1 combatants – average close to an inch taller than those who don’t make it.

Taller parents are more likely to have sons, shorter parents daughters.

I’m guessing computerisation of army records on the one hand, data warehousing and mining technologies on the other, have enabled such findings. But talk about answers begetting more questions …


2 Replies to “The returning soldier effect

  1. Ha! I do not trust this Satoshi Kanazawa. His WWI stats on taller soldiers being more likely to survive seem to me, well, questionable. (He speculates that this may be because the ratio of their vital organs to overall body size is lower – ???) Taller parents are more likely to have sons – ???

    I did a bit of searching around on t’internet (until the proliferation of nonsense out there started to give me mind-ache), and the only support for Kanazawa I could find were other references to Kanazawa. Now I’m even beginning to doubt whether the returning soldier effect, which I had hitherto simply accepted as fact, is real. I was hoping for a quick fix from a decent popular source such as Snopes or More or Less, but no luck. I did find this, but it’s not exactly conclusive:

    ‘Over 30 demographic and environmental factors have been studied for their effects on the sex ratio at birth, including family size, parental age, parental occupation, birth order, race, coital rate, hormonal treatments, exposure to environmental toxins, stress, several diseases, and war (6–9). The finding of a small but significant increase in male births during and after war has been documented in Europe and the U.S. in both the First and Second World Wars (10–12), and in the U.S. for the Korean and Vietnam Wars (13). However, studies of the Balkan Wars (14) and of the Iran–Iraq war (15) did not reproduce these findings. Proposed biological explanations for the observed increase in sex ratio during war include stress to adult males, affecting the viability of XY-bearing vs. XX-bearing sperm; changes in the age structure of the population; and higher frequency of intercourse, leading to conception earlier in the menstrual cycle, all of which have been associated with increased sex ratios in other studies (16–18). Alternatively, evolutionary explanations argue that the increase represents an adaptive equilibrium after the decimation of males during war (13), although critics argue that the increase does not last long enough to compensate for wartime casualties (19, 20). The cause of this alteration in sex ratio at birth during war remains a curiosity.’ The full article at has the citations.)

  2. Thanks Caroline. I share your scepticism on the Taller Tommy Effect. A game changer is the mix of powerful computers with the mining of huge data sets amassed by ‘our’ surveillance state and big capital. With screeds of data from customer loyalty cards, Tesco data miners see co-relationships they not only can’t explain but have little interest in so doing. If higher beer sales on Saturday nights coincide with higher nappie sales, they aren’t much inclined to ask why. They just shift the two products closer together.

    As far back as the eighties, stats man Stan Openshaw caused a stir in the Lancet. Using GIS he’d discovered correlations of clusters of certain cancers with power stations, and was now arguing that future science would be as much data driven as theory driven. I may be oversimplifying his argument since the classic model of science – observe … reflect … construct theories … derive hypotheses from said theories … text them in the observable world – is hardly bypassed by more powerful data gathering (observation)! But from a technology as opposed to scientific viewpoint, some correlations needn’t be explained to be of use. For some purposes it’s enough that they exist, as with the supermarket findings.

    It’s different for science. If A occurs at the same time as B, we want to know if A causes B, B causes A, C causes both or it’s pure coincidence. The bigger the data sets, the less likely that any given correlation is coincidental. The wars cited in Snopes are a mixed bag in this respect. America’s war on Vietnam killed some 60,000 GIs but millions of Vietnamese. If the data sources are confined to American deaths, we might – at risk of sounding callous – say the data set isn’t large enough. We might …

    Not entirely satisfactory, is it? You’d think the question big enough to warrant the kind of inquiry that would throw up a consensus: if not on the explanation, then at least on the phenomenon itself.

Leave a Reply

Your email address will not be published. Required fields are marked *