Age of puberty in girls

Why it is declining and why that matters

In a recent post, called Girls are going through puberty too early, Walt Bismarck pointed to the drastic decrease in age of puberty in girls and blamed it for the overall fear and contempt that most adult women seem to have for male sexuality nowadays. As a 25 year old woman, I can readily observe how sex negative most of my peers are. And the subject of decline in age of puberty for girls is not new to me either. While I more or less agree with Walt here, that an earlier age of puberty is related to the animus young women seem to have toward men nowadays, I think that the issue is much deeper and more complicated than he allows. In fact, if I am correct in my analysis, we might be in the middle of an unprecedented level of widespread endocrine dysfunction worldwide¹.

Addressing specifically, the age of menarche in girls- it is determined by hundreds of genetic, environmental and cultural factors, all of which are in turn affected by it. It is, in a real sense, a signifier of the felt sense of uncertainty in a society and, as abstract and vague as that sounds, I will try to explain why that is really the case here. The solution to any of the problems it causes downstream cannot be to only “correct” the signifier and ignore what it signifies. I believe that a completely new scientific framework is needed if we want to address such entangled and cyclical phenomenon in any real capacity. I see this piece as the first step toward that end, and so here is an overview of all that we actually know about the age of puberty in girls and how it actually might be causing growing hostility towards men’s sexuality .

What does the data tell us?

The average age of menarche has historically been between the ages of 14-17. The only confounding exception here is seen in the records from classical India between roughly 500BC to 500AD, that claims the age of menarche to be 12. The paper does not comment on this but I will note here that this age is very likely to be the lower end of the spectrum, and not the average, as it comes from legislative texts and not demographic data. The same texts often mention the average age of completed physical maturity in women to be 16 or older and recommend waiting at least two years from that point onward for arranging marriages.

All the historical records mentioned in the study and in its citations also show a consistent trend where, all other things being equal, wealthy, and more importantly urban, girls have an earlier age of puberty by almost 2 years. This is the reason that the study claims that the recent decline must be related to higher levels of prosperity in the modern world². But all of the historical demographic data we have available shows the average age of menarche in girls to be 14 or older.

The more recent longitudinal studies [1, 2, 3, 4, 5]³ show a clear decline in the age of menarche and thelarche throughout the world that starts roughly around the Industrial Revolution in the 19th century and seems to be proportionate to the level of both development and urbanization. There are exceptions to this trend from some developing countries that show a much steeper decline. The common hypothesis for this is that these countries tend to have much higher levels of exposure to many endocrine disrupting chemicals(EDCs) but in practice, this has only been proven true in a handful of cases.

There are hundreds of cohort studies from across the world [1, 2, 3, 4, 5, 6] that link the decline in age of menarche and thelarche to many chemical, biological, environmental and even cultural factors. It is found to be a covariate of many known EDCs along with genetic markers, nutrition (measured by various nutrient deficiencies, quality of food intake, sugar intake), health (BMI, amount of daily physical activity, participation in sports), disease prevalence(levels of various antigens), family structure(number of siblings, separation of parents, contact with parents and grandparents), cultural factors and overall life history patterns(level of education, gender ratios of household, age of marriage if applicable, number of siblings, average length of a generation).

Some of the highest correlations found above are with genetic factors. And this may seem as a contradiction to the overall argument I am trying to make- if it’s just genetics then can it even be helped by any other means? But genetic analysis often falls into this trap of an implicit assumption that genetics are more or less static in human timescales, and that they are outside of our control. The fact is that even when specific genetic markers definitively relate to behavior or biological proclivities, one has to look at population level change and wonder what is causing increased prevalence of these genetic markers in any population. Genetics certainly offer a clear mechanism for changes like the declining age of menarche, but they don’t offer an explanation of what is driving the change.

What is remarkable, is that every variable that I have mentioned above shows weak but highly consistent correlation with age of puberty- meaning that if you know the estimated correlation between some variable and age of menarche, you can predict the exact extent of its effect down to a precision of a few weeks, and yet you can‘t explain the entire population level decline by just combining the effect of all the known covariate variables. Clearly, there is an additive effect of all the identified factors where, when combined, they have an effect that is much greater than the sum of its parts.

It should also be noted that for many of the correlations, the path of action is either unknown or reversed, meaning that covariation can mean that some factor affects the age of menarche or the age of menarche affects that factor, or frequently, both. The best example of this is adiposity: the research shows that adiposity of a very specific pattern is relatively highly correlated with early menarche- but this kind of adiposity precedes the actual onset of puberty. This implies that the hormonal pattern that causes adiposity is the same one that causes early puberty but that the extent of its effect is regulated by the external factor of calorie intake.

Studies done in the US, Greece, Spain and Denmark [1, 2, 3, 4] specifically find a change in what is termed the ”menstrual tempo” where the correlation between the start and end of puberty, which was historically quite significant, seems to have come unhinged- the age of thelarche seems to have decreased a lot more than age of menarche(but both have decreased). The reason for this is very likely to be a higher exposure to phytoestrogens and other estrogen inducing pollutants in early adolescence, as the onset of thelarche happens through estrogen receptors, while onset of menarche happens through the hypothalamic GnRH mechanism that has a greater genetic component to it. In fact, the age of thelarche can likely be used as a benchmark for direct exposure to EDCs, while age of menarche is instead highly sensitive to a generalized sense of stress through the hypothalamic mechanism.

This data includes generational changes by including the parents/guardians of the adolescent girls that are the subject of the studies. This means that the steepest decline observed goes as far back as the 1950s, coinciding with quite an astounding number of cultural, economic, and lifestyle changes in America and much of the world. Changes like the widespread use of hormonal birth control, of formula instead of breastfeeding, of many consumer products that use EDCs in their manufacturing, of “instant relief” drugs like DES and thalidomide that treat symptoms and not causes of various maladies etc. Or perhaps the economics of meaningless fiat currency, the push for inefficient global markets, the filter of bureaucracy on every economic activity etc. Or taking another perspective, the many “end of history” narratives of the last half century, increase in atheism and nihilism, a drastic increase in divorce rates and crime rates, changes in law and perception of abortion, suddenly and drastically increased rates of immigration etc.

I think that it is possible to actually go into quite some detail and figure out the mechanism by which each of these changes has an effect on something like age of thelarche and menarche by linking them to the myriad correlations mentioned earlier. I will leave that exercise for the readers. However, even if the readers disagree about the specifics here, I think we can all agree that all of these changes have contributed to the overall perception of instability in everyone’s minds. The hypothalamic mechanism is quite sensitive to perceptions, whatever the actual reality might be, and so I think we can conclude that it is definitely reacting to these factors, either directly or indirectly.

Lastly, one of the most prominent environmental factors in age of puberty, that I want to highlight here is father absence [1, 2]. Out of all the environmental factors already mentioned, the current levels of fatherlessness in the developed world are the most novel and unnatural for humans⁴. Historically, the only condition that has ever really warranted supplanting of fathers from their children’s lives to such a degree is active warfare, and yet we seem to be artificially mimicking this condition in our modern hubs of prosperity. It cannot be just a coincidence that every young person is seemingly in shell shock despite living in objectively one of the safest eras in history⁵. Every evolutionary biological account of human behavior presents paternal involvement and investment in their children as one of the most salient innovation contributing to the exceptional successes of humans as a species. Are we really ready to just abandon it now as something superfluous?

Life history theory

It is quite clear that there are too many factors at play when it comes to age of menarche, and their interrelations are not always easy to sort through. Moreover, in such a nested system, even when proximate cause and effect is determined, there is no way to determine what is actually driving the overall change. It becomes an exercise of trying to map a coastline by examining one fistful of sand at a time.

One possible solution to this lies in the life history (LH) theory. It is a theory that relates various ecological and resource constraints to biological/life strategies of survival- specifically, individual vs group fitness. According to it, all biological organisms have specific patterns of behavior to optimize either individual or group survival based on their overall ecological environment. For example, in a harsh but predictable ecology, where causes of mortality are predictable, it is much more effective to invest in individual organisms that would have a higher chance or survival. Whereas, in a more abundant but unstable ecology, where causes of mortality are unpredictable, it is more effective to focus on group survival over individuals. Of course, reality is never dichotomous and every environment presents a unique set of circumstances and variables. LH theory maps the ways in which we adapt to them. This does not only involve conscious and deliberate behaviors, but often primarily concern with the unconscious and neurological/semi-autonomous biological processes, that nonetheless quite effectively react to the environmental cues.

The Verhulst model⁶ is a commonly used logistics model, for population growth with an upper bound for resources⁷, and it is often used as a basic framework for understanding LH strategies. It describes the patterns of population growth for different biological strategies on a spectrum of quality vs quantity of offspring- with the assumption of limited resources, it becomes necessary to prioritize one over the other. On the r-K spectrum that it uses, humans as a species sit on the far K end- meaning that all humans by far favor the quality of offspring more than the quantity. However, within human populations, there is variation in r selected or K selected behavior.

This model explains why, seemingly contrasting conditions can still produce the same outcome in life history strategies. For example, the highly stable and unchanging rural conditions, either historical or present, as well as the consistently dangerous conditions of chronic illness or war, both delay the age of menarche, favoring K selection. While, the unpredictability of suddenly improved nutrition or high calorie intake as well as that from the separation of one’s parents, both reduce the age of menarche, favoring r selection. LH strategies do not change linearly with the amount of resources or danger in the environment, they instead optimize for the specific level of harshness and unpredictability of the ecology.

Ultimately, our physiology, endocrinology and genetics seem to converge to decide which kind of life strategy is best adopted in any environment⁸. For example, this meta analysis specifically explains how lower paternal investment might be working as a signal to trigger a faster LH strategy in adolescent girls and therefore causing the age of menarche to decline. The implication here is that an earlier menarche is a strategic change that would result in greater number of years of fertility and therefore greater chance for having more offspring. Consistent with this theory, girls that experience an earlier menarche seem to also start having regular ovulatory periods in a shorter amount of time since menarche.

If the change toward earlier menarche is really driven by a shift in LH, then here are some of the cues and signals of unpredictability and instability that might be causing it:

• Fatherlessness and lack of other close family relations that might provide a feeling of safety.

• Overall cultural sensitivity amplifying feelings of unpredictability that are unfounded.

• Higher refined sugar intake, which has historically been an extremely rare resource, as well as all the foods that are much more calorie dense than they used to be.

• Various nutrient deficiencies, like magnesium and vitamin D, that seem to increase with industrialization and urbanization.

• Rise in many illnesses from a diminished gut microbiota⁹

• The presence of many EDCs in the environment, that create false signals and cause many forms of hormonal disruptions.

Why the sex negativity?

Even though the LH theory and the Verhulst model help explain the declined age of menarche, they don’t explain the sex negative feelings of aversion and hostility that young women seem to have developed for men’s sexuality. A faster LH is, after all, generally associated with a higher socio-sexual behavior and not lower. So what is really happening here?

I believe that a big part of the sex negativity is fueled by the earlier onset of thelarche, that is caused not by just a change in LH strategy but by the specific estrogenic chemical pollutants. Thelarche might be defined as the budding of breast tissue, but it is a complex biochemical process that goes along with the development of other kinds of secondary sexual characteristics, such as various facial characteristics, shape and positioning of hips, length of eyelashes, body hair, redness and clarity of skin in various areas and various body odors. The earlier onset of all these changes brings with it sexual attention that is inappropriate and extremely overwhelming and distressing at such a young age. As Walt has pointed out, this very likely creates at least a mild trauma response to any sexual attention, even when it is in a perfectly appropriate context.

Secondly, the earlier menarche and faster LH form as a survival and, more importantly, reproductive strategy. While in it, a disproportionate amount of resources are invested into faster reproduction- in a sense, the whole point of a faster LH is to have more babies, faster- the actual fertility rate in young women is currently incredibly low. Despite the many reasons or means by which this is the case, it works as an artificial signal of temporary resource scarcity or famished conditions for our endocrine systems and therefore fuels an aversion to anything sexual.

To add to this effect, most commonly used forms of birth control work by stopping ovulation, which is something that only occurs naturally during either pregnancy, or again, famished conditions. Which means that the women that are on these forms of brith control, or are exposed to the residual progestin in our water supply- essentially mimicking the same effect, are actually signaling to their neural and endocrine systems that they are not fit for reproduction, one way or another. It is not much of a surprise then, that they have a visceral negative reaction to male sexuality.

Conclusion

We seem to have fundamentally altered our ecology in the modern world to such a degree that our biology is now adapting to it instead of adapting to any of the conditions outside of the artificial- despite all the varied natural environments worldwide, just based on the age of menarche, we are all clearly converging to the same, faster LH strategy. We have seemingly created many artificial cues to promote this strategy- which inevitably accompanies group fitness over individual fitness, a lower investment per child, and lowered individual output in every sense. And within the faster LH domain, we are creating strong artificial signals of resource scarcity.

In a way, I see most of the cultural, political and social issues of our time as the direct result of this fundamental change of ecology. If we are to deal with any of the problems it has caused in any real capacity, we have to understand the underlying phenomena- treat the illnesses and not just its symptoms.

1. Along with the declined age of puberty for girls we are also seeing a catastrophic decline in men’s levels of testosterone, along with all the negative health outcomes it implies, a dramatic increase in every kind of illness that is caused by endocrine disruption, like hypothyroidism, metabolic syndrome, higher adiposity, respiratory problems, PCOS, endometriosis, various fertility issues, low sperm counts, increased rates of certain cancers etc. It is impossible not to see all these negative trends as a cluster of changes that are caused by the same or similar events.

2. There is also a trend where colder climates and higher latitudes correspond with a higher age of menarche by just as big a margin. Wouldn’t it be just as legitimate to claim that the recent decline is due to global warming?

3. I have tried to take out any studies here that are not available to read either for free or through a common academic institutional login. Apologies, if any of them are still inaccessible. However, there really are hundreds of longitudinal cohort studies that show this trend and they should be quite easy to find anyway.

4. We can quibble about the historical quantity and quality of involvement of fathers in the lives of their children here- there are definitely many accounts that go against the perhaps rosy picture that most people paint of the past. But anyone pointing to any historical instances of widespread fatherlessness might want to consider whether it presents as an ideal or even acceptable context for them to live in.

5. Despite the recent increases in crime rates that I have mentioned earlier, on a longer timescale the crime rates and especially the severity and ubiquity or violence in general today is lowest in recorded history. We must not forget that the historical accounts for the age of puberty cited earlier are mainly from the time periods where public executions, gruesome corporal punishments and frequent religious persecution trials were the norm.

6. As with any mathematical model that describes any real phenomenon, one has to be aware of its shortcomings. Mathematical models are usually quite good at describing the skeleton of a system but aren’t particularly effective at the minutia of its internal processes. This model specifically has been a popular meme by the name of “r-K selection” theory. I enjoy the memes as much as the next man but it doesn’t hurt to get to know the underlying math when possible.

7. This is not the only accepted model for population growth, and has both add-on or complimentary models as well as competing ones. But it is still a good heuristic to understand patterns of behavior.

8. The EDCs in this case are only acting to increase the felt instability and stress to reinforce the broader cultural or evolutionary-behavioral changes.

9. Note that gut microbiota is directly responsible for the synthesis of many neurotransmitters and therefore can directly change neural signaling.