Hacker News new | comments | show | ask | jobs | submit login
Humans evolved to tolerate smoke poisoning [pdf] (gwern.net)
70 points by gwern 34 days ago | hide | past | web | 46 comments | favorite



Smoke still is bad for humans though, see this essay: https://www.samharris.org/blog/item/the-fireplace-delusion/

>There is no amount of wood smoke that is good to breathe. It is at least as bad for you as cigarette smoke, and probably much worse. (One study found it to be 30 times more potent a carcinogen.) The smoke from an ordinary wood fire contains hundreds of compounds known to be carcinogenic, mutagenic, teratogenic, and irritating to the respiratory system. Most of the particles generated by burning wood are smaller than one micron—a size believed to be most damaging to our lungs. In fact, these particles are so fine that they can evade our mucociliary defenses and travel directly into the bloodstream, posing a risk to the heart. Particles this size also resist gravitational settling, remaining airborne for weeks at a time.

>Once they have exited your chimney, the toxic gases (e.g. benzene) and particles that make up smoke freely pass back into your home and into the homes of others. (Research shows that nearly 70 percent of chimney smoke reenters nearby buildings.) Children who live in homes with active fireplaces or woodstoves, or in areas where wood burning is common, suffer a higher incidence of asthma, cough, bronchitis, nocturnal awakening, and compromised lung function. Among adults, wood burning is associated with more-frequent emergency room visits and hospital admissions for respiratory illness, along with increased mortality from heart attacks. The inhalation of wood smoke, even at relatively low levels, alters pulmonary immune function, leading to a greater susceptibility to colds, flus, and other respiratory infections. All these effects are borne disproportionately by children and the elderly.

>In the developing world, the burning of solid fuel in the home is a genuine scourge, second only to poor sanitation as an environmental health risk. In 2000, the World Health Organization estimated that it caused nearly 2 million premature deaths each year—considerably more than were caused by traffic accidents.


I'm sure there is some truth to the article, but it paints a very different picture than the referenced paper at the buttom.

> Finally, returning to the questions posed at the start, we conclude that although there is a large and growing body of evidence linking exposure to wood/biomass smoke itself with both acute and chronic illness, there is insufficient evidence at present to support regulating it separately from its individual components, especially fine particulate matter. In addition, there is insufficient evidence at present to conclude that woodsmoke particles are significantly less or more damaging to health than general ambient fine particles.

> Nevertheless, given the importance of woodsmoke as a contributor to particle concentrations in many locations, strategies to reduce woodsmoke emissions may be an effective means of lowering particle exposures. In addition, given the weight of toxicologic evidence, additional epidemiologic studies are needed to confirm our conclusions.


Wow, thanks for following through there.

As much as I complain that sloppy research oversells its conclusions, I tend to forget how often reporters will take a good paper and simply skip all the lines like "not significant" and "no causation" that stand in the way of an exciting article.


>In 2000, the World Health Organization estimated that it caused nearly 2 million premature deaths each year—considerably more than were caused by traffic accidents.

Everyone is rallying around autonomous cars to decrease unnecessary deaths (and rightfully so), but relatively few folks are familiar with rocket stove technology.

https://richsoil.com/rocket-stove-mass-heater.jsp

https://en.wikipedia.org/wiki/Rocket_stove

Releasing smoke is energy inefficient, since it's literally unburnt fuel! Considering the fuel savings, the lifetime cost of switching can be negative in some cases.

In terms of cost per statistical life saved, it's an extremely compelling technology.


It's the same with a mass oven. It has several burn chambers that is designed exactly yo burn all smoke. Heck, I even have a portable stove that uses secondary combustion to burn all the wood and smoke. If I ever have to buy a wood stove, I want one which is designed to burn it all. I probably even go all in with a mass oven.


Thanks for sharing those links. It looks like it would be hard to start the rocket action correctly. You would have to preheat the circulator "chimney" somehow. Until then, smoke would pour into your house.

This is a problem with normal wood-burning cooking stoves as well. Smoke doesn't go into the chimney until you establish a convection draft or the wind creates a venturi effect.


Thanks! Great points.

Smoke free startup is an important design/operational consideration. There are lots of videos demonstrating startup technique (search "rocket stove starting"). Here's a couple good ones: https://www.youtube.com/watch?v=Ls_eqNB6d0Y&t=1m23s https://www.youtube.com/watch?v=aSkUquOQt5w

Note that the goal of the drafting chimney is not to soak up a lot of heat, but rather to be insulated and have low thermal mass to minimize heat loss. The higher the air exit temperature from the riser, the stronger the draft. The important thing for self-drafting is to heat the air in the riser, and only secondarily heat the riser material itself.

A good design should make startup as "idiot proof" as possible. I've seen a few builds that use a 12V case fan on the air intake. Since the goal is to heat up the air in the riser (and that heating the riser... draws hot air into the riser), all you have to do is "tip it over" into that stable equilibrium.

Imo there's a market opportunity for an inexpensive, compact, nice looking, and easy to operate rocket stove. The open source (and excellently documented) "batch rocket" design seems a good place to start. http://batchrocket.eu/en/


Solar hot water heaters are vastly more convenient and generally cheap. But, rocket stoves are great for rarely used structures, cooking, or the far north.


Yeah, solar thermal is woefully underused. Even something like attached south-facing greenhouses and solar hot air panels can cheaply retrofit passive solar design onto the existing building stock.

To that list of rocket stove applications & advantages would add backup heat for solar hot water, accessible to developing countries and the poor, and (when built from masonry) ease of construction in the charred afterscape. I'm kidding of course on that last one. Mostly. ;)


Of course it's not good

But if your options are, stay close to the fire or die, that's an important selective pressure


> There is no amount of wood smoke that is good to breathe. It is at least as bad for you as cigarette smoke, and probably much worse.

And yet very few sane people wish to ban wood smoke. It smells good. There's nothing like the smell of fall and winter, with fragrant smoke blanketing the countryside.

I'm perfectly aware that wood smoke is bad for me, that it pollutes my lungs and makes my food poisonous. But it's delicious.


I don't smoke, but I am a beekeeper, so I am often breathing in smoke. I've always worried a little if that was bad for me. I guess I will try harder not to breathe it in. Without it, I would get many more stings, which I assume is also not good. There is a chemical replacement for smoke, but that is probably even worse.


Try telling that to my idiot hippie neighbors who drive a prius and have weekly bonfires.


Why not tell them yourself? Passive aggression on a news feed won't improve the situation.

On the other hand, if the neighbors are burning wet wood, pine or otherwise Doing It Wrong maybe they would appreciate a friendly heads up.


Why not help them build a Dakota fire pit?

It's essentially just a big hole with an air intake tunnel, so that fires in the pit combust more completely, with a very significant reduction in smoke and particulates.

You probably can't do anything about them being ideologically inconsistent with respect to burned fuels as an energy source, but you can reduce their impact on yourself and your mutual neighbors.


A bonfire's smoke moves in the wind's direction. If they sit towards the bonfire with the wind in their backs, what's wrong with that? Although if the smoke goes toward your home I understand you anger.


Yeah, this seems not-terrible - most of the serious health concerns with fires are with wood stoves, where smoke either enters a house directly or leaves a chimney and settles back into it.

From the article up there I get the sense that wood fires inflict a lot of particulates even when you can't smell smoke, but open-air fires are still way safer (on this level) than in-home ones.


Prius (virtue signaling) vs. bonfire (actual behavior).


The idea that Priuses are only about virtue signaling is silly. For people who don't much care about power and handling, they're good cars that can be bought for a reasonable price and are cheap to run.


I don't buy the smoke poisoning angle.

They compared with Neandertal AHRs and found that human AHRs may have undergone selective pressures since they're quite different, so the authors tried to connect these changes with the invention of fire. However AHRs bind to many things.

Also, Neandertals also used fire extensively (see for example http://www.pnas.org/content/108/13/5209 ) so they should have undergone the same selective pressures, so they should have the same or similar changes as humans?


Might the selection pressure have been fire related, without necessarily being just the use of fire?

If, for one speculative example, Neanderthals mostly used fire outdoors, while modern humans used it indoors with correspondingly thicker smoke concentrations, wouldn't that account for the difference? Or perhaps only a few Neanderthals out of any given population used it regularly (for crafting tools, say), while almost all modern humans could have used it for cooking food.


I see! But that's a bit hard to prove, isn't it? I don't think we have enough archaeological sites with proven fire places to prove that assumption


" Our findings reveal that a functionally significant change in the AHR occurred uniquely in humans, relative to other primates, that would attenuate the response to many environmental pollutants, including chemicals present in smoke from fire use during cooking. "

So we have some genetic tolerance to bad stuff created from cooking? Cool.

This does, however, raise ever more questions about studying the effects of chemicals on rats and primates and drawing conclusions from that to human populations.


This seems like a really good question to raise, and I'm embarrassed that I didn't think of it at all. I know carcinogen studies are already touchy since cancer risks vary so wildly, but this is a nice reminder that even within primates you get this kind of problem.


Reminds me of a story where Robert Sapolsky, a professor in biology when he was staying in a native African hut and decided to put out the indoor fire before sleeping, since he was concerned about the smoke. Later during the night he woke up, covered in driver ants, realizing very fast the reason why the natives sleep with the huts full of smoke.


This is interesting. I wonder if the tolerance of the mouth to heat (relative to other areas with a thin epidermis) is also related to cooking.


I've often wondered whether something similar might be at play in relation to the supposed cancer risks from acrylamides forming in burnt food. Sure, the stuff's poisonous to rats etc, but given our physiology has evolved smaller intestinal tracts and larger brains in relation to cooked foods (particularly, meat), might it also be that we're less susceptible to its cancerous potential than other mammals?


Lung cancer?


The way the title is phrased is strange. To the young/uneducated reader (not so many around here, but still), it reads like humans evolve to develop a particular trait, when we know that people who developed that trait got an evolutionary advantage and ended up spreading their DNA more than others on average.

Not as catchy a title, sure, but at least, it doesn't misrepresent the evolutionary process.

NOTE: I am not an evolutionary biologist, so there is most likely an even better way to rephrase my rephrasing. I'd be glad for anyone to do so, the more accurate, the better.


> humans evolve to develop a particular trait

> people who developed that trait got an evolutionary advantage and ended up spreading their DNA more than others on average.

The first is shorthand for the second.


It's sloppy shorthand though because it can be (deliberately) misconstrued to suggest agency and directional purpose.


Ah OK, my bad then.


> it reads like humans evolve to develop a particular trait, when we know that people who developed that trait got an

people who developed that trait had to evolve to do so, no? Not as an entire population, but maybe over the live span of individuals.

Maybe the smoke itself caused that, as exposure often does, "evolving" in some individuals who then pass immunity markers down over generations through blood in the womb rather than genetically. Or maybe the smoke is cancerous and caused beneficial mutations. I should read the paper (edit: it says genetic). All I'm trying to say is, either development is subsumed under evolution.


Evolution occurs via reproduction over the span of many generations. Individual adaptation on a single-lifespan scale occurs via epigenetic methylation tagging of the existing (evolved) genome.

You cannot become better adapted to smoke unless your genome already contains a gene that mitigates the damage, or a gene that can regulate expression of a damage-mitigating gene based on that specific stimulus.

So if you breathe smoke frequently over your lifespan, you may have a gene related to mucus in the lungs expressing its protein at maximum capacity 24-7, while someone who never breathes smoke has that gene "turned off" most of the time.

Transfer of individual adaptations, usually in the form of antibodies that can teach immune cells to respond to a specific hazard, is very limited, and will only last as long as that specific threat is continuously endemic. We have mostly supplanted that natural capacity with artificial vaccinations.

Random mutations are usually harmful. Sexual reproduction and selection is the mechanism to shuffle the most harmful mutations out of the genome while still retaining the rare beneficial mutation. If the mutation rate is too low, the species cannot evolve quickly enough in changing environments. If the mutation rate is too high, sexual reproduction cannot eliminate the bad mutations fast enough.


> people who developed that trait had to evolve to do so, no?

The word "evolve" in English means a gradual change, so no.

Random genetic mutation isn't a response to the smoke, and there isn't anybody born unadapted who then adapts by exposure. The species evolves, individuals don't.


>"The species evolves, individuals don't."

You are made of cells that are many generations separated from the fertilized egg. I don't see why you accept evolution at the species level but not cell level.

This sounds like a kind of reverse creationism, where macro-evolution is ok but micro-evolution is not.


Um, honestly not sure what you're getting at there. The individual cells in your body are genetically the same as what you started with, with occasional random mutations. The body defends itself against those mutations and most will be eliminated. Some will become cancers. They don't affect the DNA you pass on to your offspring and so aren't a part of the evolution of your species. None of them are going to give you superpowers unless you happen to be living in a comic book.


Say a person is exposed to smoke often since being a young child. At some point, a lung cell is dividing and randomly gains a mutation that increases expression of some smoke-protective protein. This lung cell is then more fit (to the smokey environment) than the surrounding lung cells. As a result, its daughter cells will divide more often than the others and populate the tissue.

Are you denying this can happen? Or is it that you do not consider that process to be evolution?


And then that person has kids, and obviously they don't get the mutation. Is that part of the process of human evolution? No.

(By the way, you didn't mention the most common way that lung cells mutate in a smoky environment and begin to divide more often.)

Evolution involves random mutations and the filter of natural selection. The process you're describing doesn't participate in the filter, and random mutation alone gets you vastly more downside than upside. The only reason evolution works is because of the filter which lets us keep the upside in the gene pool while the downsides are filtered out.


That absolutely can happen - but remember that the traits of the 'evolved' lung cell will not be passed to offspring, as the genetic makeup of gametes (sperm) is a random (50%) subset of the main genetic code.


Sure, but the population of cells we call an individual human being does "evolve" over the course of its life.


The vast majority of random mutations are harmful.

The number of random mutations necessary to produce one lung cell that is better adapted to smoke inhalation is large enough that the entire remainder of the lung already has carcinogenic mutations.

Sexual reproduction is a mechanism to eliminate harmful mutations without discarding the occasional beneficial mutation.

It is not necessary for that purpose, but it is sufficient. Humans are now technically capable of identifying a potentially beneficial genetic mutation, and inserting it into an unrelated person's genome. It is, however, more likely that the protein result of the mutation will be used to research and develop a pharmaceutical drug or medical procedure that replicates the beneficial effects of the mutation.

In short, the probability of you spontaneously developing an adaptive mutation in a non-germline cell is absolutely infinitesimal in comparison to the probability that someone in the last 10000 generations of your ancestors developed such a mutation when producing a sperm or egg cell and passed it along to you.


Many interesting points here that I wish were backed up by refs.

> "The vast majority of random mutations are harmful."

- You sure they aren't "neutral"?

> "The number of random mutations necessary to produce one lung cell that is better adapted to smoke inhalation is large enough that the entire remainder of the lung already has carcinogenic mutations."

- No, in principle it can be a single point mutation to a promoter that increases affinity for a transcription factor.

> "Sexual reproduction is a mechanism to eliminate harmful mutations without discarding the occasional beneficial mutation."

- So are apotosis and immune surveillance.

> "Humans are now technically capable of identifying a potentially beneficial genetic mutation, and inserting it into an unrelated person's genome."

- No, they aren't. They modify and select cells in a dish, not in a person.

> "It is, however, more likely that the protein result of the mutation will be used to research and develop a pharmaceutical drug or medical procedure that replicates the beneficial effects of the mutation."

- Ok, I am just including this for completeness sake.

> "In short, the probability of you spontaneously developing an adaptive mutation in a non-germline cell is absolutely infinitesimal in comparison to the probability that someone in the last 10000 generations of your ancestors developed such a mutation when producing a sperm or egg cell and passed it along to you."

- You are talking of 10^4 fertilized eggs vs 10^13 cells in my body[1]. I'm not sure how many divisions it is from egg to egg, but it has to be less than the 10^16 supposed to happen during the human life span[2].

Actually I am pretty sure I have read somewhere that the human body probably contains at least one cell that has a mutation at any given bp. I will see if there is time to find it later.

[1] https://www.ncbi.nlm.nih.gov/pubmed/23829164

[2] http://bionumbers.hms.harvard.edu/bionumber.aspx?id=100379


You forgot to include the probability that any given mutation occurs in a cell where such mutation can be observed to produce a beneficial effect.

If one cell in your body mutates such that it might protect your lungs against damage from smoke inhalation, it won't do you any good (nor ill, probably) if that cell is in your toenail matrix.

Nor do all cells in the body divide at a uniform rate.

I would guess that the one type of cell most likely to show an effect from non-germline mutation are the hematopoietic stem cells. They produce all types of blood cell, are self-renewing, and blood cells touch every other part of the body. But they are also only 1/10000th of all myeloid cells. There may be only 10^6 HSCs in one body

Your reference [1] only included the abstract: https://www.researchgate.net/publication/248399628_An_estima...

Also, 10000 generations of ancestors is not 10^4 fertilized eggs. Most people have two parents. 0 generations of ancestors is 1 fertilized egg. 1 generation is 3. 2 generations is 7. 3 generations is 15. The upper limit on N generations is 2^(N+1)-1 fertilized eggs. For 10000 generations, that is 4e3010 cells. Though the number of individuals in the species eventually limits that number, as multiple inheritance causes some of your ancestors to be counted more than once. The lower limit would be the degenerate case where all 10000 generations were siblings that had at least one son and one daughter, and that is 2e4. So quite a lot of room for variability in there.


>"Nor do all cells in the body divide at a uniform rate."

Knowing that rate in the various tissues even to within a few orders of magnitude is the holy grail.


I think you are saying that there is no teleology in the theory of evolution.




Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | DMCA | Apply to YC | Contact

Search: