News and views about philosophy, the academic profession, academic freedom, intellectual culture, and other topics. The world’s most popular philosophy blog, since 2003.
Cyber security professional here -reliably determining when a computational artifact (file, etc.) was created is *hard*. This is sorta why…
Agreed with the other commentator. It is extremely unlikely that Pangram’s success is due to its cheating by reading metadata.
I see this question as a bit naïve. There is metadata on every document created by a modern word processor…
There’s a simple way to test. Open a pre-2022 essay and copy-and-paste it into a new file.
At the risk of self-advertising:… You claim “AI is unusual in degree, not in kind” and “It is not clear…
Apropos of Sagar’s wish to foist the A.I. industry by its own petard, this article appeared in print in yesterday’s…
I teach both large courses, like Jurisprudence and Critical Legal Thinking (a.k.a Legal Argumentation), and small seminar-based courses at Edinburgh…
This is an astonishing finding. Any biologists or philosophers of biology care to interpret its meaning for us?
This post suggests the study is over-interpreted and more rigorous investigations are required http://epgntxeinstein.tumblr.com/post/127416455028/over-interpreted-epigenetics-study-of-the-week
This is not really so surprising. Environmental effects on gene expression ("epigenetic" effects) are well-known, and there are many well-established examples of their inheritance across generations. It remains controversial whether such effects can last over many generations, and this doubt enables conservative evolutionists to keep the spectre of Lamarck at bay. Given that environmental causes of epigenetic effects may themselves be stable, and they may even be stable because the organisms they affect make them so ("niche construction"), the relevance of the alleged decay of epigenetic inheritance over many generations to defeating Lamarckism is unclear. It may still be argued that Lamarckian responses to the environment must have evolved by traditional Darwinian processes.
Epigenetic effects have been theorised about and observed for at least two decades now, but this is more or less the first clear case found in humans.
And yes, the most prominent epigentic theorist, Eva Jablonka – who I think holds a position in a philosophy department at Tel Aviv – has described herself as a "neo-Lamarckian".
What’s going on here? Briefly: DNA is not the only biologically significant thing we inherit from our parents. Most obviously, perhaps, we typically inherit an environment – a culture, climate, diet, … – which interacts with the genome (and much else besides) to produce a phenotype. So, two populations might vary in height because of genetic differences; but they might be genetically identical and vary in height for nutritional reasons. The supposed disparity between the average North Korean and South Korean is a useful example. Similarly, anyone of contemporary average height who has had to duck on entering a medieval building will know that populations have become much taller since then, without there having been time for sufficient genetic change to explain it.
Now, since environments are substantially shared, we don't often see sharp discontinuities from one individual to the next that we can attribute exclusively to environmental differences. The Korean border is a very unusual case in this respect; differences in height between my family and the family who grow up next door are probably mostly down to genetics. But consider the uterine environment. That's crucial to a significant part of an infant's development, but is typically not shared by neighbours and the like. So a more acidic womb environment, say, might lead to differences in height. Moreover, it could be heritable; suppose that being conceived in a more acidic womb also makes one likely to *have* a more acidic womb. Then we would expect to see shorter parents have shorter children and taller parents have taller children, *even if* all members of the population were genetically identical clones, and received the same diet, exercise, etc.
Now suppose that a population's food supply is extremely variable; some years there is lots of food, some years there is little. And suppose that there are two developmental pathways available to each embryo; Bigchild and Littlechild. When sufficient nutrition is available, it's better to follow the Bigchild pathway; to grow bigger. When insufficient nutrition is available, it's better to follow the more economical Littlechild path. So the population might evolve a developmental sensitivity to the mother's diet; if the mother is well-fed, her children will follow the Bigchild pathway, if she is undernourished, they will follow the Littlechild one. If her diet is in the normal range, they will as before follow the same pathway that she did. In that case, we might see what looks like a Lamarckian evolutionary process; a period of famine results in the birth of smaller children, but they too have smaller rather than larger children for several generations afterwards.
Likewise, glossing over a lot of biochemistry, we know that the cellular environment is important for processes of gene activity and expression. The same genetic material can be activated in different ways and at different times, resulting in different phenotypic outcomes. And this too is known to be heritable, meaning that we can get inherited variation in the absence of genetic, uterine, or more conventionally "environmental" differences. We call these epigenetic effects.
What seems to be happening in the Holocaust survivors' case is more or less analogous to the Lamarckian story sketched above. The extreme stresses engendered by Holocaust exposure, unsurprisingly, alter the parents' cellular functioning. But since parents pass on to their children not just DNA but also the cellular functioning which activates it, the children too show altered genetic activity, despite the absence of genetic change. This would be extremely useful where children encounter a similarly traumatic environment to their parents’; in effect, the body “bets on” encountering the same environmental stresses as the parents encountered, and can respond to them in advance.
This (Guardian) post put me in mind of the following blog post by David Spiegelhalter (Winton Professor of the Public Understanding of Risk at the Uni of Cambridge):
http://understandinguncertainty.org/heuristic-sorting-science-stories-news-0
tl;dr: "Ask yourself: if the study had come up with a negative result, would I be hearing about it? If NO, then don't bother to read or listen to the story."
This is not an attack on journalists or newspapers, or those who read them. Heuristics are not perfect. But it speaks to the effect of publication bias and how much greater it can be when a story goes from journal to national newspaper.
So – is this a new dawn for biology? Or has a rare outcome within the bounds of randomness occurred? I don't like to cast aspersions without doing my homework but I tried to read the article and one has to pay for it. The abstract gives virtually no information about methodology and zero statistical results apart from a single mention of a significant correlation.
Dr. Glackin – that's a really fascinating post, thanks for taking the time to write that all up. I'm particularly interested in the point you made about cellular environment/functioning inheritance. How exactly do parents pass on that cellular functioning to children if not through their DNA? Is it something only the mother can do, and it gets "passed along" based on conditions the child experiences as it develops in the womb, or is there another means by which that cellular functioning gets passed down to the next generation?
Apart from the compelling and more basic concerns about the possible mechanisms whereby the so-called epigenetic effect might operate, consider the N sizes in the holocaust survivor studies:
"Cytosine methylation within the gene encoding for FK506-binding-protein-5 (FKBP5) was measured in Holocaust survivors (n=32), their adult offspring (n=22), and respective demographically comparable parent (n=8) – offspring (n=9) controls."
http://www.biologicalpsychiatryjournal.com/article/S0006-3223%2815%2900652-6/abstract
Most eye popping is the pathetically low n for the controls, 9 and 8, although 32 and 22 for the experimental groups is pathetic enough.
It's pretty clear that the lead author of this study has been banging on the purported epigenetic connection between holocaust survivors and their offspring for many years. Obviously, if you investigate enough genes with small enough samples, you're eventually going to find some result that is "statistically significant". There are just tons of genes to pick from in all such genetic studies — that's what makes them so unreliable unless careful methodology is strictly followed.
But, of course, the ideological commitments of so many people strongly favor such results. Who's going to look that gift horse in the mouth?
"How exactly do parents pass on that cellular functioning to children if not through their DNA?"
First, when cells divide they pass on a lot of cytoplasm with many chemicals and independently reproducing structures in addition to DNA. Second 'through their DNA' is ambiguous. Mostly people seem to mean through nucleotide sequence. But chromatin structure and other kinds of epigenetic marking are also inherited, which are factors that largely determine gene expression. Much of this is clearly heritable, or liver cells would not generally give rise to liver cells, muscle cells to muscle cells, etc. Much of this epigenetic structure is deleted in the formation of gametes, which has led to a tendency to think that trans-genertional epigenetic inheritance is problematic or improbable. We now know, however, that a good deal of epigenetic information is inherited. Disputes are about how much and for how long. The fact that the environment can induce epigenetic marking makes it difficult to determine how much is directly inherited, though a properly broad account of inheritance would include some environment induction, at least when the parental organism creates the environmental conditions which give rise to the epigenetic changes. In the holocaust case it seems fairly clear that the inheritance is direct, i.e. the chemical structure of the chromosome reproduces itself in some relevant non-sequence related respect.
Here are a couple of articles expressing the skeptical — and widely held — view among scientists regarding so-called epigenetic effects:
Thanks to Lexington for pointing out the very small sample size (from which we are wondering if it's safe to infer to the population of all humans (around 6 or 7 billion) and then to humans that don't yet exist!).
As soon as I saw this I went to the principle of charity and thought, 'Hmm, well maybe the study had good statistical power to detect the effect of interest'. But, statistical power is presumably behind the paywall. So no opinion either way on that matter.
[Out of interest, am I the only one troubled at the reference to some scientists electrocuting mice repeatedly just to see if they could then repeatedly induce 'fear' into them and many more mice by other means? Not sure how that one got through ethics. Reminds me of the study done electrocuting dogs through the floor, to get them to jump a barrier to safety, and then electrocuting that area too, just to see how long a dog will jump before collapsing from exhaustion.]
Jerry Coyne thinks the study was problematic: https://whyevolutionistrue.wordpress.com/2015/08/24/holocaust-trauma-is-it-epigenetically-inherited/
By calling epigenetic effects "so-called", I take it you are implying they are not considered scientifically valid. But read the Ptashne articles you cite carefully and you'll see that what Mark is arguing against is not the existence of epigenetic effects but rather the mislabeling of certain phenomena. To quote the FASEB article: "There is a form of gene regulation that is properly called epigenetic (17): an environmental signal can cause a gene to become transcribed (by a DNA-binding activator), and transcription of that gene can be maintained in the absence of the original signal. The effect can be perpetuated over many cell generations and can extend to fully differentiated cells (8, 18, 19)."
It is this proper sense of epigenetic effects that is being examined in the article that prompted this post. (Shane Glackin's explanation is very good.) And people reading this should be interested to learn that the "trans-generational transmission of trauma" has been very well documented by psychologists and psychiatrists for decades–one of the gifts of WWI and WWII to science, I suppose — but there has not been a clear mechanism to account for the remarkable case descriptions. Epigenetic effects would indeed be the most reasonable candidate.
I don't know anything about this particular study (did not read) but yes, Lamarck was right in some cases. Jerry Coyne says there is not "any known case" of multigenerational epigenetic inheritance. Uh, that is totally false. For just one example, Sally Mackenzie's lab here at University of Nebraska has shown multigenerational epigenetic signaling in a variety of plant species, and importantly I only know this because she is a floor above me, not because I know all of the cases. It is astonishing someone claims to know all of the cases. Mainstream journalism and the "skeptical" blogosphere have somehow succeeded in presenting science as far more unified than it is…biology is immensely complicated and most biochemists (for example) are only intimately familiar with the enormous complications of the few genes and proteins they are working on. To publish a short opinion article (whether in a journal or on your blog that advertises for your popular books) and present it as a scientific consensus is a bit much. So I do not think these views are "widely held" among scientists, though they may be widely held among science journalists and bloggers (but the matter is empirical so I should leave it there).
If it sounds like the issue grinds my gears it does! I do not think Lamarck was right in every case or that there is only Lamarckian inheritance. But clearly the idea that he might have been right in some cases (and the jury is still out…so what side should we err on?) has major implications for our personal responsibility that the Weismann/Morgan/Mendel dogma does not. Shane Glackin (#3) makes a good comment about the issue.
Responding to VL, I don't know how anyone can read the articles I linked to by Ptshane and find any comfort for the view that so-called epigenetic effects (I so call them "so-called" because from occasion to occasion the term is applied to wildly different phenomena) really can support inheritance of acquired characteristics. Yes, he mainly focuses on how histone modifications can't be inherited — as is assumed by some who talk about "epigenetic effects" — but he says nothing that would suggest that other mechanisms in human beings might do so. The most advanced organism for which he describes an inheritable mechanism is a nematode, which uses bits of RNA in the sperm to pass on immunity to a virus. But to do so, he observes, it uses an enzymatic process human beings don't possess.
And this is the larger problem with the notion that human beings (or for that matter, more advanced animals) might participate in this wonderful Lamarckian evolution: there's no established mechanism whereby the transmission, along the germ line, of inherited characteristics might take place; it is pure speculation that such a mechanism exists.
Plant biologist says that he is aware of epigenetic transmission in the case of plants. But plants aren't animals, and there's no reason to think that the mechanisms found in them might apply to animals. What gets through and may work in the germ line of plants may be very different in general from what gets through the germ line of animals.
I think that the comment of Shane Glackin demonstrates some of the problems in discussions of "epigenetic effects". For example, he presents at some length the case of a mother affecting the outcome of her children through an acidic womb. But what does this have to do with "epigenetic effects"? How are the effects he describes not understood in the usual way as simply the effects of prenatal environment, which has been studied extensively, and no one imagined to suggest a kind of Lamarckian evolution? "Epigenetic effects" has become something of a grab bag in which all manner of effects that affect both parents and their children are said to be "inherited" by some mechanism built on genes.
Finally, I encountered another useful commentary on the Holocaust survivors study from a scientist at the Center for Epigenomics at the Albert Einstein College of Medicine. Beyond noting the obvious statistical weaknesses of the study, one of the more telling points he makes is this:
"Without wishing to beat this study to death, the fact that the DNA methylation difference that characterises the exposed parents is an increase while that characterising the offspring is a decrease is difficult to reconcile with any sort of heritability mechanism."
The findings on holocaust second generation stress reported is interesting but not so shocking. there is a lot of epigenesis around. The well established cases are all the result of adding methyl groups to the outside of DNA molecules that block the expression of certain genes–including perhaps in this case genes that code for neurally expressed ones (40 % of all genes are expressed in the brain). The key thing to note is that the methylation of DNA bases is itself controlled by other DNA sequences, regulatory genes that control the expression of structural ones.
FKBP5 methylation is affected by a variety of environmental factors eg social class in a much larger study. The question is whether the (inverse) correlation between parent and child here might be due to fetal environment (fetal programming hypothesis), early childhood environment, correlation between current environments, parental treatment of the child, diet or many other factors.
Leave a Reply to VL Cancel reply