Thursday, June 23, 2016

The end of the "population boom": the Seneca collapse of Ireland's population during the great famine

The story of the Great Famine in Ireland is a starkly clear example of a "Seneca Collapse", that is of a case where decline is much more rapid than growth. Is something in store for us at the global level? (Image source Rannpháirtí anaithnid (old) at English Wikipedia)

In a previous post, I argued that many current global population projections are mistakenly based on the idea that the "demographic transition" will work backward. That is, it is often assumed that impoverished people will tend to make more children and that, therefore, the world's population will keep growing even in the midst of the profound economic decline that could accompany a resource and climate crisis. (this is, for instance, the assumption of the original "The Limits to Growth study in 1972)

I proposed, instead, that the start of a major economic/climate crisis will cause an immediate reduction of the birth rates in part as the result of the declining health of fertile women and, in part, by a rational response by families who would understand that they can only care for a limited number of children in a condition of increasing poverty.

That is, there won't be a population increase in the midst of a major crisis and the decline of birth rates would immediately bring the start of a worldwide population decline that could even take the shape of a true "Seneca Collapse." To support my argument, I brought the example of the Soviet Union, whose population started declining even before the political collapse of the Union. I also mentioned several examples of other Western Countries (e.g. Italy) where birth rates have been going down in parallel with the worsening of the economic conditions, to the point that we are starting to see an overall population decline.

This interpretation was criticized in the comments by some who argued that, yes, my interpretation may be correct for relatively modern and "Westernized" countries, but not for the poorest areas such Africa or Asia. These commentators argued that people in these areas will continue to make as many children as they can, no matter what happens around them; apparently as a result of the Imams telling them to do so (or by evil dictator Erdogan, or the like).

I don't think this criticism is correct and I can counter it with an example. We all know the story of the Irish famine that took place between 1845 and 1852, and that killed a large fraction of the Irish population. We know something about the number of deaths, about how many Irish emigrated, but we know relatively little of how the famine affected birth rates. Did Irish women try to compensate for the higher mortality by having more children?

On this point, I found a comprehensive study produced by Phelim P. Boyle and Cormac O Grada on Demography, Vol. 23, No. 4 (Nov., 1986), pp. 543-562 (here is the link). It takes some assumptions and extrapolations to determine the Irish birth rates before and after the famine. They also do not report graphs, but only tables. However, their conclusion is clear: birth rates declined with the famine in Ireland. In other words, the Irish families didn't try to compensate for the higher mortality by having more children; not at all.

This is confirmed by what we see of the Irish population in the decades after the famine. Even though food supply ceased to be a problem, still the population continued to decline or remain stable well into the 20th century. The Irish of those times didn't have good contraceptives, but they seem to have coped mainly by retarding the marriage age and by adopting a lifestyle that discouraged sexual activity among young people.

This is relevant for the case I am discussing here. In the 19th century, the Irish peasants were Catholic (or, if you prefer, "Papists"). The catholic view of marriage was supposed to be at that time (as it is still today in some circles) that a married couple should have as many children as the Lord sends them - that is as many as possible. After the famine, the Irish remained Catholic, but they totally disregarded the advice that the may perhaps have received from their priests. It was a perfectly rational choice - the Irish were not stupid.

We are clearly discussing something difficult to quantify, but I tend to think that most people on this planet are not so stupid as PR specialists seem to believe. So, from these historical examples (Russia and Ireland) I would say that a future economic/climate crisis will be immediately accompanied by a decline in birth rates and, hence, of population. If that happens, it will be a good thing as the pressure on the ecosystem will be reduced. That doesn't mean that all the problems we face will be solved, but at least we don't have to worry that people will worsen the situation by breeding like rabbits.


  1. I don't think the immediate impact of the collapse is the contentious issues here -- I in fact agree that there will be both an increase in death rates and a decrease in birth rates during the worst years of it.

    However, what will emerge on the other side is more important.

    And that is indeed very likely to be the demographic transition run backwards combined with a reversal of secularization.

    Ireland is irrelevant to that question, because what emerged on the other side of the famine was an industrializing and secularizing society, a continuing trend until today in fact.

  2. But that's exactly the point: a model based on the demographic transition that goes backwards says that the birthrate of an impoverishing population should increase. It doesn't.

    About Ireland being irrelevant, I think it is not the case: in Italy during the same period we had also a secularizing and industrializing society, but population kept increasing rapidly. But we didn't have the famine.

  3. Again, I think we're talking about different time scales.

    The most likely scenario I envision is the following (excluding a full-scale nuclear war):

    There will be a bottleneck period during which birth rates will go down together with the increase in death rates. And it will probably be quite non-uniform in different places.

    I have no idea how long that will last, may be a few decades, may be a significantly longer than a century.

    But whatever emerges after that will most likely be an agrarian patriarchical society (pretty much doomed to never industrialize to the same level again), and also one that is a lot more religious than what we have now. And whatever religions survived/emerged after that will have probably been quite radicalized in the process, and I don't expect them at all to have internalized any deep scientific insights about what caused the collapse in the first place.

    Which is precisely a reversal of the demographic transition combined with a return with a vengeance of religious fundamentalism.

    And then the overshoot-and-collapse cycle restarts once again.

    You are talking about a temporary period of declining fertility, but on a sufficiently grand scale that is indeed irrelevant, the question is can we meaningfully escape out of the overshoot-and-collapse cycle. And I have very good reasons to be pessimistic about that.

    1. I think that over the time scales you are mentioning, human society will adapt and will no more be bumping all the time against the limits of the ecosystem. Religions, well, I think they are a transient phase of humankind's history. Think that nothing like what we call today a "church" existed before about two thousand years ago. And last three centuries - more or less - saw a remarkable decline of the organized forms of religion that we call churches. I think it is evolution at work. The future will see completely different religious forms

  4. Loosing a quarter of its people to emigration (close to half those who were in their reproductive years) influenced significantly the shape the population curve of Ireland after 1840.
    Now, suppose that immigration comes for decades to Italy or Ireland at the same rate. This is tomorrow's challenge isn't?
    To really grasp the total picture, please consider that Norway, 125 years ago, had less people than those of Norvegian origin living in the USA.
    Why? Because renewable energy cannot maintain dense populations where winter is cold.
    So, when oil gas and coal will be past their peak, the only option left will be to get numbers back to what they were circa 1910. The sooner we understand this, the better.
    Hortense Michaud-Lalanne

  5. First time commenter here. I like your blog, but in this case I have to ask.
    Can you explain what numbers/graph exactly provide a "starkly clear example of the Seneca collapse"? Do you mean the figure at the top?

    It struck me because just yesterday I read a very interesting story about the potato and how Ireland had about 30 different famines in the hundred years leading up to the "Great famine".

    I'm asking because looking at 50 years before and after the peak, I don't see anything that resembles "slow growth/fast collapse". It's fast growth, fast collapse.
    Looking at the entire period, I see overall long term growth.

    Can you explain where the Seneca collapse is? Since the overall growth in this picture seems unperturbed, I'm having trouble to call it a collapse even.

    Thanks, Frank

    1. In this case, the "Seneca" shape is not so clear as in others. Overall, however, if you look at the curve you see that it is asymmetric; it does go down faster than it grew, at least in the decade of the famine. Then, there are many kinds of collapses and so is life

  6. Just to add, I kind of understand the point you are trying to make. Catastrophe, famine, "collapse" does not equal sudden increase in birth rates. But I would be careful with using Ireland as an example, because it didn't actually suffer a collapse, if incomes and lifestyle continued to increase shortly after the famine.

    1. Yes, Ireland is just one of the examples. If you look at the previous post, you see that I brought the example of the Soviet Union and of other countries. My thesis is that, generally, people tend to manage the number of their children depending on how they perceive the future. There seem to be several cases supporting this hypothesis, but, of course, there may also be cases supporting the opposite. As usual, we try to extract "laws" out of the real world, but sometimes it is difficult

  7. But what caused the huge increase from 1780 to 1840? The reason for the collapse is obvious, but if you ignore the increase and the subsequent collapse, you could just about fit a straight line between 1600 and 2000, in other words steady growth.

    1. see
      and of course the bones used
      Hortense Michaud-Lalanne

  8. Ugo there was a paper by the late Professor Alf Bartlett on population growth and decline which while taking a different approach in presentation agreed with your views and what the data shows. Bartlett however made a distinction about what he called the time lag between between the onset of factors that make people have less children and when this actually occurs which puzzled him as to why it was not immediate so to speak. I no longer have the paper but my recollection is that the time lag may be up to 70 years and was tied to issues of fertility or fecundity. The proposition made sense to me and may explain why people cannot see why there is no change immediately.

    1. That would be an interesting paper to find. But he wrote many things....

    2. Ugo. Bartlett was a master of the exponential function and derivatives but he was especially interested in applying this mathematical theorem to the concept of growth and demonstrating by simple mathematics that exponential growth of anything you like was always going to end. He prepared a paper on coal in the country in which I live and demonstrated mathematically that given known resources if depletion was allowed to continue in line with growth then those very same resources would last at best 50 years (and that was some 15 years ago already).

      Population growth and decline has many factors to include within the any such function. I had reason to research American native indian populations pre conquest and post conquest, the outcome from the available data (and it was not accurate but reasoned guesstimates was that the native American Indian populations fell of a cliff but the factor at play was actually disease and it led to a population crash similar and probably worse than the Black Plague in Europe in the middle ages. So disease is a factor in decline as well as resource access or loss of resources, so is violent conflict another factor. I guess I raise this as within the broader systems outline in the Club of Rome work I thought population was too simplistic. For instant; Dr James Lovelock asserts that the human population will crash to less than 300 million by 2100, that is a very steep curve given the current world population. It is a fascinating subject because it is uncomfortable for us to consider rationally the demise or loss of our fellow human beings on a very large scale and especially to consider what that really means.



Ugo Bardi is a member of the Club of Rome and the author of "Extracted: how the quest for mineral resources is plundering the Planet" (Chelsea Green 2014)