Some people really really don't like seeing images of spiders, and so I would recommend hiding it behind a like, spoiler tag. Or, if that's not possible, remove it and just have a link to it.
I really really really don't like seeing people advocating accommodation of every little issue people have. How do you suggest we accommodate my problem?
Can it really be considered manners or empathy when the best available evidence suggests that the vast majority of trigger warnings don't accomplish anything?
Well, that's that. I bow before the preponderance of evidence and, going forward, will not tag any pro-accommodation-of-others advocacy with content warnings, because it will not accomplish anything.
More seriously, none of the studies included in the meta-study examined the usefulness of specific warnings against common phobias to phobic populations. With the exception of Bruce & Roberts (2020), they examined general/trauma-naive populations and generic warnings, of the form:
"TRIGGER WARNING: The passage you are about to read contains disturbing content and may trigger an anxiety response, especially in those who have a history of trauma"
"The video could..."trigger extreme distress among some people, especially survivors of trauma."
"Warning: This study involves viewing photographs that show emotional events. Some may be very graphic and very negative in nature (e.g., trauma, war, torture, maltreatment and death). Some people may find this material distressing. Please do not proceed if you do not want to be exposed to this material or think that you may be adversely affected by being exposed to this material."
“Warning: This study involves recalling a negative personal experience. Some people find this process distressing. For example, you may experience negative mood and intrusive mental images. A small minority of people also experience distressing memories and reactions in the week after recalling negative events, although these reactions generally subside quickly. Please do not proceed if you do not want to take part in this task or think that you may be adversely affected by this task.”
Et cetera. Not quite the same as "cw: arachnophobia" to avoid a spider.
It's not a great surprise that such warnings did very little for the psychic comfort of general population undergrads and MTurkers, and may well have been anxiety-inducing themselves.
Conversely, it's painfully obvious to me that if someone doesn't want to view a specific kind of image in a post delivered to their mailbox by Substack, providing an option of avoiding it is useful to them. And yes, consideration of the feelings of others is the core of good manners - hell, even Emily Post would have recognised the definition.
"The vast majority of trigger warnings accomplish nothing" is sadly entirely compatible with "CW spider" being very useful.
In this case, the spider doesn't trigger me because it doesn't fit the schema of my phobia, but if it had, I would have been very grateful for the TW, as it'd avoid me breaking out into sweat in the middle of the sweltering summer heat.
Make it easier to make a comment private so only the author can see it. It didn't need to be a public comment. I feel annoyed at people pointing out typos, which is another good use of author-visible-only comments.
In regards to the accommodation, my wife is arachnophobic and having a surprise large spider picture appear on the blog she loves to read (and which gets auto-delivered to her email inbox) would really upset her. Fortunately I got to it first and warned her not to read this one. So yeah, I'm not just advocating in the theoretical. I felt like I'd managed to save a loved one from unpleasantness and thought perhaps I could help others.
Phobias are best dealt with through exposure therapy, I believe, so in the long run, you're not doing her any favors trying to insulate her from a world where spiders, like, exist and stuff.
> Psychotherapists treat arachnophobia with exposure therapy, too. They expose people first to cute, little spiders behind a glass cage. Then bigger spiders. Then they take them out of the cage. Finally, in a carefully controlled environment with their very supportive therapist standing by, they make people experience their worst fear, like having a big tarantula crawl all over them. It usually works pretty well.
> Finding an arachnophobic person, and throwing a bucket full of tarantulas at them while shouting “I’M HELPING! I’M HELPING!” works less well.
(The post also has general discussion related to the topic at hand.)
Since we're on the internet and without the benefit of verbal tone, I'll say explicitly that this is not meant to be confrontational-- but this seems like an odd ask for an article about bug cognition. Lots of invertebrates can give humans the heeby jeebies, including bees themselves. It does not sound reasonable to request animal pics be hidden behind spoiler tags because someone might get scared.
I disagree with this. People who don't like looking at pictures of insects should look at this as a chance to get a bit of exposure therapy which will help reduce their anxiety response.
By having pictures, this article is doing those people a great service. They just don't realize it. Instead they are being given a scary "content warning", implying that looking at such pictures may be harmful, which is the opposite of the truth.
So, the content warning here is harmful, the pictures are not, and may be helpful when viewed in the right way. There is some research to back this up:
(Note: if something is very disturbing and unpleasant to a majority of people then yes, by all means hide it, there's no need to make everyone's day slightly worse. But for things that most people don't have a problem with but a few people find very triggering, I think what I said above applies. This is not discrimination against those people, it's actually helping them recover from their trauma / get rid of their annoying anxiety response.)
Exposure therapy is done thoughtfully and gradually It involves teaching people how to calm themselves and working up from small challenges to larger ones.
It is *not* throwing difficult stuff at people and telling them it's a favor.
For the sake of potential discussion, a tanget: "Exposure therapy" while I expect is often helpful, sounds a lot like rebranded meditation of the last 6K years.
Well, there ARE forms of meditation that use the same principle, but definitely not all of them. Some formal Buddhist meditation seeks to remove the fear of death in various ways, several of which use something that could be called "exposure therapy", but that's not exactly the same thing as treating a phobia. And I said "some" because there are very different branches of Buddhist meditation, and they don't all do this.
That is admittedly true. But it seems the slow gradation is more to keep patients from leaving therapy not because its necessary to do it that way. It seems teaching relaxation techniques helps speed the process along but its not clear if its completely necessary.
Getting triggered, while unpleasant, should generally be helpful in the long term, unless the person is conceptualizing getting triggered as harmful, which is the idea that content warnings promote. (I'm assuming most people viewing this page are doing so from the comfort of their home.. that should help)
There's a therapeutic approach that uses maximum intensity exposure - it's called flooding
I feel like the image of the spider was gratuitous, and also that most people I know well enough to have a confident opinion about, myself included, have the "don't like seeing spiders" response to a pretty high level. This is a pretty small sample, but I thought it was worth raising anyway.
I have no problem with spiders. I think most spiders are cute. OTOH, I'm mortally afraid of horses. And cute unicorns with their deadly horns induce panic attacks in me. Also, men with beards freak me out. And penguins scare the shit out of me. I've had dreams where thousands of penguins attack me and peck me to death. So please don't show me pictures of horses, unicorns, men with beards, or penguins. Please respect my triggers. Thank you.
Wouldn't it be useful if AI to provide us with predictive trigger warnings by looking at people's comment history and profiles? That way I could be sure I was upsetting the right people in the right way! ;-)
I suspect you are being facetious / trolling and don't actually want an honest response, but just in case you would like one, here's a copy/paste of the response I made to a similar comment made up above:
"My suggestion for those who dislike seeing trigger-warning-requests: Make it easier to make a comment private so only the author can see it. It didn't need to be a public comment. I feel annoyed at people pointing out typos, which is another good use of author-visible-only comments.
In regards to the accommodation, my wife is arachnophobic and having a surprise large spider picture appear on the blog she loves to read (and which gets auto-delivered to her email inbox) would really upset her. Fortunately I got to it first and warned her not to read this one. So yeah, I'm not just advocating in the theoretical. I felt like I'd managed to save a loved one from unpleasantness and thought perhaps I could help others."
I wasn't serious about asking you to remove your comment, but neither was I trolling; I was making a point. While fear of spider pictures is irrational (and presumably your wife agrees), the chilling effect of asking people to self-censor every single thing that might upset anyone is very much real. I wasn't lying when I said I really, really dislike such requests. It is not possible to accommodate everyone's sensitivities, and it is not anyone's duty to try. I do, however, see it as my duty to push back on such requests so that some semblance of free speech is still possible.
I don't know your wife, so I'll talk about a generic arachnophobic person instead. The first time the person sees a picture of a spider, she will be upset. Will she be equally upset the second time, the third time, and the 100th time? Even single-celled organisms exhibit habituation. It would be odd if arachnophobic people don't. On the other hand, if they're "protected" from ever seeing a spider or even a picture of a spider for their entire lives, their sub-conscious never learns that (most) spiders are harmless and substitutes an ever-scarier villain for the mundane reality.
I live with someone with arachnophobia, so I kinda get it, but... this was a really noncentral example of a spider photo, in an article about insects. It's not black, doesn't have many legs visible, doesn't set off the "blob with a bunch of radiating lines" danger-sense. This is roughly equivalent to seeing a photo of Spider-Man in an article titled "Top 10 Superheroes"
> Does that mean bees feel emotions? If they feel emotions, would that mean bees have conscious states?
I see no reason to doubt that bees feel emotions and are conscious. Obviously, I can't prove that, but I can't prove that about my fellow humans, either. It simply seems to be the best explanation.
Consciousness is orthogonal to intelligence, which I would measure operationally.
Well, one rub is that "having" consciousness likely isn't binary, it's probably a continuum. Where along the continuum, in turn, dictates how we should treat them if you tie your ethics in with consciousness. Should we murder 100 bees for 1 human? 1000?
Plus it is theoretically possible they are not; i think it's reasonable to assume plants are not conscious, or at least have so little of it that it's basically nothing.
Interesting! I would love to hear more about this perspective. To me, I do not ascribe any moral weight to rocks very specifically because rocks are not conscious. If rocks gained consciousness, I would suddenly find myself ascribing them moral weight. It is hard for me to avoid the conclusion that my ascription of moral weight seems to be caused entirely by consciousness.
How would you handle that? Do you ascribe moral weight to rocks now? Would you ascribe the moral weight if they were conscious?
"Ascribing moral weight" isn't a primitive in how I think about ethics.
I eat many mammals, which I absolutely believe are conscious to a high degree, although not as intelligent as humans. This is not because I do or do not "ascribe moral weight" to them. I think it's very important how we treat animals, and plants as well.
I think it's also important how we treat rocks, although I wouldn't treat them like animals. I don't know how I would treat rocks differently if I thought they were more conscious than I do.
I see ethics as structured and mediated through particular relations, and the relation of "being of the same species" is an important one, among many others. I do not believe in a generic "moral weight," as a utilitarian might.
I do not quite know what a moral weight is. The phrase by itself is meaningless. I guess it is a shorthand for a number of factors, but probably different people think of different factors when they say "moral weight". For me one of the important factors is that I do not want to cause suffering, and humans are capable of much more richer suffering than lower animals because of their complex awareness. Of course this cannot be the whole story (see e.g., all those problems in ethics about killing a person in deep coma). I accept eating meat if the animal was killed without suffering (in best case without even knowing it was going to die), but I would not accept treating humans in the same way. So there is something more I cannot yet fully express or justify. Or it Is simply time to admit that vegans have a point.
I, as well as almost every other human on Earth, do not eat other humans for three reasons:
1. Humans are an extremely inefficient food source, whether they're hunted or farmed.
2. Humans do not taste good.
3. Most humans seem to be naturally averse to harming other humans without reason, as well as the act of eating their flesh.
In a world where none of those facts were true, we would probably have human factory farms in most countries. If cannibalism was something that was just a normal part of life, people would find justifications for it, as humans always do. It has nothing to do with any complex morality.
Same reason I don't eat zebra; it's too hard to come by. Likewise, same reason most animals don't eat humans; they taste bad, and also they shoot you when you try.
Humans - lower animals - plants in descending order of organic functioning in the ecosystem/mystic connectedness/natural order.
The inanimate(e.g. rocks) are purely resources to be used in rhythm with the natural order and out of consideration of the needs of other organics. This describes the fundamental moral struggle of self vs. self in society([archway engraved w/ 42?]) of the conscious, I'd say.
That's a very traditional hierarchy, and it has a lot of tradition going for it. But it's not a valid argument, though it indicates that a valid argument is likely available.
I agree that there are different roles in the ecosystem. I wasn't suggesting homogeneity. I was just noting that the way we use natural resources is also important.
Stephano Mancuso might have an issue with your assumption that plants have "lower organic function." The average tree has more rootlets than a human has neurons, and they communicate with each other via at least 15 different chemical communications systems. ;-)
I think I actually assign vastly more moral weight to rocks than animals. My view on factory farming is "yeah, I'd like the conditions of animals to be improved as long as it doesn't affect my burger's price". On the other hand, I quite frequently become sad when I drive on a road cut through rock, as I ponder the permanency of the scar that'll remain long after the road is gone.
Factory farming, of course, having no effect at all upon the landscape or environment.
I can't imagine the worldview that values purely arbitrary aesthetic considerations — if not these, but instead e.g. as in an environmentalist framework, then the importance of such things lies entirely in their effect on the biosphere ("save the rainforest, xxx species die every x minutes!", not "save the rainforest, we like pretty rocks!") and hence your position would be nonsense on the face of it anyway — ...
...— purely arbitrary aesthetic considerations above living beings / suffering / conscious beings / biodiversity and biosphere / etc.
Like, I don't think I've ever actually seen even the edgiest teenager try to claim "I don't care about anything except how things look 🩸☠️". It's sort of impressive, almost, in a sad sort of way.
There is also a possibility, though this is rather a very vague idea, that consciousness is like computation power: once you have certain minimal set of abilities you have all the abilities (like simple Turing machines being able of any possible computation), though, perhaps, in a slower package.
If your definition is like mine, not quite. Consicousness is essentially a sensory function, not an implementive function, so consciousness in and of itself can't enable anything. You need to combine it with an action mechanism. Just knowing that your toe hurts doesn't do anything for you (say you've stubbed your bare foot), you've got to wash it and bandage it. The sensation, the consciousness, lets you know that you need to do that.
Similarly a Turing machine can do any possible computation, but it can't fix your breakfast of smell the smoke in the air from the burnt toast. It's complete within a restricted domain.
It is possible you are right. However, self-consciousness may be a necessary side effect of certain level of integration of brain functions, and the ability to self-model (sorry for being vague). So I do not think consciousness can be separated from "action" functions of the brain.
What would it even mean to have emotions without consciousness? You can have automatic danger avoidance reactions, its., but I thought emotions presuppose some kind of conscious experience
Another thing that I would associate with consciousness is the "play" behavior. Like with everything else it is easy to misinterpret, and caution is warranted not to leap to conclusions. And here is some recent research which made headlines about bumblebees playing with balls (apparently young bumblebees were more playful):
I see no reason to doubt that bees feel emotions, and they are definitely intelligent to at least some degree. Whether they're conscious or not is going to depend on your precise definition of consciousness. They meet my definition, but I tend to shade in the direction of panpsychism, i.e. I consider an air conditioner to be minimally conscious. (If you disagree, it's probably because we differ on exactly what "conscious" means rather than disagreeing about anything substantive.)
I always thought that before we try emulating human intelligence, we should try studying and emulating insects. They are far, far simpler (and hence potentially easier to understand and emulate than higher animals) but still capable of astonishingly complex behavior. And in many practical applications insect intelligence would be more than sufficient. Think of self driving cars or autonomous delivery systems: putting an equivalent of a bee brain to steer your car would be amazing.
I am not do sure for the self driving car example: small animals seem agile, but it's mostly because of their immense strenght to weight ratio and body resistance, and also speed of mouvements that hide the details to human vision: when looked at in macro and slowed down, fast insects like bees or ants looks clumsy and bump in many obstacles, just rebounding.
Big animals like a human or, even worse, elephant, move much more accurately and carefully cause they operate in a regime where bumps or imbalance result in injuries. Cars are much more like the later, so would not really be a good fit for fast, approximate "let's try until success" control strategy.
A very good observation, I might have been too carried by my enthusiasm. But in any case, I did not mean copying exactly bee brain to pilot your car, and there are ways to overcome what you pointed out above: for example, you do not want cars to bump against one another as a part of your movement algorithm, but having 360deg LIDAR permits you to bump virtually, which is quite acceptable. Secondly, roads are normally quite predictable, and you can adapt insect-like algorithms to utilize this predictability and give equally predictable reactions. Note that bees live in a very complex, unpredictable environment, and have to move very fast in 3d: simple randomized approaches may work better in this case than more complex intelligence. When things go bad on the road (some accident), the simple but fast insect-like brain might come in handy, and perform better than more sophisticated approaches. Of course it may happen that in case of self-driving cars you really need high-resolution computer vision in which case this application of insect brains is off, but from engineering perspective, insect cognition which does so much with so little is very attractive.
Small(ish) NN similar to insect brains are certainly powerful, and would offer good preformances especially if hard-coded in specialized processors. Afaik self driving cars are already using this approach at least partially, they use NN trained on stimulated and actual roads.
My impression is that self driving is mostly solved in the sense it works in the vast majority of case, probably as well or better as the average human driver.
But it's not 100%, and I don't think anyone is willing to take responsability for the rare cases it fails. Constructors don't, passengers neither (who want to be passenger with the legal responsabilities of a driver? A passenger that should be able to take back control at any time, so can not do anything but passively watch the road? I don't ).
so it's kind of an ill posed problem, and even the case where everybody (including myself) thought the cake was cooked (road freight, trucks and such) is in fact not clear: humans do a lot more than driving (Paperwork, legal responsability, deterrent to stealing,....) and their salary is not huge. Driverless trucks are not able to compete when you take that into account.
Moving from A to B with a car is imho already solved by current tech... But it seems human drivers do a lot more, and even if they kind of suck to safely go from A to B, making them simple passengers proved hard in practice.
By the way, the critical issue of responsability for accidents makes NN or other similar techniques inherently problematic: for classic programming (procedural or variants) you can trace (in principle) the reason for some action to specific part of the algorithm, maybe even actual lines of code. The system is deterministic and can be analysed using classic reductionism (Mostly. Bugs and poorly written software already makes this very hard sometimes), and this is very convenient to establish responsabilities...
NN are still determinist (or could be, even if i think many include randomness because it often improve outputs) but can not really be analysed in the same reductionist way. So what if it fails? A bug? A exceptional circumstance that could not have been anticipated? Insufficient training? Who pay for it? Who get fired?
This is why there is some interest in neuro-symbolic/explainable AI: it still uses NN, but with a structure which is (sort of) understandable to humans: In some cases it means that NN encodes a sentence/set of sentences in fuzzy logic (it may be a very large set of sentences, and the logic is, well, fuzzy, but it is still better than conventional NN in terms of analyzing what AI does. I am not sure there are any attempts to apply neurosymbolic computation to self-driving cars, but it is certainly applicable in others (like expert systems judging people, where you can actually discover a bias). My best guess is that the future robotic systems (like self-driving vehicles) will have some instinctive component based on traditional NN (maybe inspired by social insect brains) and some symbolic component encoding moral, legal, and physical rules and getting veto.
The city of San Francisco doubts that it's as competent as the average driver. It tends to get into stalls in bus lanes and won't get out of the way of emergency vehicles.
I guess eventually yes, especially in case of space alien who created sufficiently advanced civilization, though some victims of initial misunderstanding are probably inevitable. Which, unfortunately applies also to aliens visiting the Earth, and that is one more reason not to abandon technical progress and get to them before they get to us
'I have often wondered about alien life. If we ever met one, would we recognize a space alien as "intelligent"'...
The recent advances in AI made me think about this as well. I'm no longer so convinced that humans are even the smartest creatures on this planet... if there is some living thing that's much smarter than us (a massive fungus brain?), how would we even know? I don't think dogs have a very clear understanding that we're "smarter" than them. If something out there is doing inscrutable hyperintelligent things, people probably wouldn't even notice.
More seriously, how do you define "intelligent". If you're going to say that we aren't the most intelligent creatures on the planet, you need a definition that can be tested against. Clearly most of the ones we use are human-centric, but what else would you propose?
I think I have a fairly general definition that would work: intelligence is the ability to identify and make use of similarities between different things, at increasing levels of abstraction.
Like, geometry starts when you notice that a round stone and a round tree trunk have something in common. Management/psychology start when you notice that different people react in similar ways to you whacking them, and you use this to obtain compliance from more people. And the ability to recognise that the marks-on-paper I love you are the same as the waves-in-air je t’adore is several levels of abstraction higher, and indicates more intelligence.
I don't really think that would work, unless you consider all patterns detected equally valuable. Noticing that a tree branch has a circular cross section is a lot easier for a person than for a rhinoceros. We've got grasping members specifically designed to grab small cylindrical objects which we inherited from our tree-dwelling ancestors. An octopus would be more likely to notice how well a sucker adhered to it, or what the surface color variation was...which you didn't even consider significant enough to mention. (And all patterns would have to include echos and UV coloring and, oh, smells, etc. Those are what some species consider important patterns.)
"...consider all patterns detected equally valuable."
In terms of detection, I think we do have to say that, because we have no way to prejudge which patterns are going to turn out to be valuable, and which aren't. But my definition does say "identify and make use of" - so if some individuals or some group seem to identify useless patterns, and others seem to identify useful patterns and make use of them, then we could conclude that the latter group is more intelligent.
As for the other variations you mention... a truly general definition has to work across all of those difference. If octopuses are very good at observing sucker-related patterns, and on that basis develop a sucker-based mathematics, then we can safely say that they're intelligent.
I realize this isn't a problem for everyone, but the style of this was very distracting. I could practically feel the unnecessary unrelated meme pics every third paragraph break. And the unexpected spider prominently on display immediately when loading the page, while relevant I guess, didn't exactly give me a warm fuzzy feeling to start reading with.
It's understandable if you found the style distracting, but I'm confused what you mean by "unnecessary unrelated meme pics". Aside from the spider at the top, aren't all the pictures directly related to what's being described in the text? And (IMO at least) none of them are memes
There exists a style of blog posts, in which the flow of text is frequently interrupted by "meme pics" which are related to the content of the preceeding paragraph or sentence, and which are supposedly intended to lighten up the mood of the reader. This style was somewhat popular a couple of years ago and has thankfully fallen out of fashion now.
"Maybe later" meant that they were reminded of that blogging style.
The prose is right on the edge, for me. I very much respect the Feynmanesque decision to keep the essay conversational and jargon-free, but a few of the puns might have been better left off.
The subject matter is fascinating, though, and I did pick up the book.
In general, I agree. I think most of the community reviews could use a bit more of this type of humor, but this one in particular used too much. It actually got in the way of my comprehension.
Yeah, I had to stop for awhile and figure out the Theabee line - I couldn't figure out what part of the experimental setup that was, and took a bit to realize it was a joke.
A bit easier after that, after I adjusted to the style, but a bit distracting at first.
I have read several of the great bee books (like Honeybee Democracy), and the topic gets better and more remarkable as one moves along. Thank you for this review. I loved it. I suspect that most readers will stop in the middle, but for those who continue, the end material is the best of all.
Humans are already quite predictable. Otherwise complex societies would be impossible. Of course quite predictable does not imply completely predictable, but neither are insects.
Yeah. I found that statement very odd. Humans are quite predictable and we predict other humans and their reactions, in aggregate or at an individual level, quite accurately, just not perfectly.
There’s no way we could safely operate tons of metal at 60 mph if we weren’t, in at least some conditions, extremely predictable. Air travel emphasizes this in a different way - you wouldn’t get in a metal tube with 300 strangers at 30,000 feet for six hours if you didn’t predict they would behave basically the way they actually do.
It would just help people realize that consciousness is an incoherent concept, and that we're already able to answer for any given things whether it contains any of the real subcomponents of consciousness, like ability to learn.
About moral weight? First, science alone cannot answer value questions. Moral weight depends on your ethical assumptions. But of course science can (potentially) answer if those assumptions are satisfied in particular cases. In this case, sometimes it seems we can and we do. The research underlying the book "Consciousness And The Brain" in a last year's entry into book review contest:
is apparently useful as a factor in deciding the fate of people in deep coma. It can also be used to debunk the claims of Peter Singer about morality of killing born babies (they are probably conscious, contrary to the claims by P.Singer).
So, as pointed out in the aforementioned review (as well as many other posts on this blog) we already know or suspect quite a bit about the role of consciousness and its underlying neural mechanisms. Of course we are no closer than ever to solving the "hard" problem of consciousness because it is probably unsolvable. I find the following analogy as useful when thinking about hard problem of consciousness: Once the question about the true nature of matter was an important part of philosophy. Then came physics, and now we know the structure of matter to the teeny tiny bits, and about the forces which hold it together. But did it solve the real "hard" problem of the nature of matter? I am no philosopher, but I guess the answer is no. However, aside from some fanatic philosopher here and there nobody cares anymore: our knowledge about how the matter operates, and how to manipulate it, and curiosity about what we do not know yet in this regard displaced almost completely the interest in the "hard" problem of matter. I believe the same will happen with the "hard" problem of consciousness: it will die natural death when our knowledge about practicalities of consciousness become deep enough.
“I believe the same will happen with the "hard" problem of consciousness: it will die natural death when our knowledge about practicalities of consciousness become deep enough.”
This is exactly the argument of Anil Seth in his book Being You.
He describes what he calls the “real problem” of consciousness, which is “to explain, predict, and control the phenomenological properties of conscious experience.”
In "What It Is Like To Be A Bat" by Thomas Nagel, proposes that we can never know what it is like to be a bat, because we didn't develop our mental capacities as bats. To know what it is like, you have to actually be a bat. Interestingly, I believe that same thing, more or less, is true for individual people.
Yep. Or rather, whatever it is will come up with a new name for what we humans do, and will reserve "consciousness" for whatever ineffable processes are exclusive to itself.
Hard to ask if things are conscious when no one can define consciousness in a coherent purely materialist way. Just like free will, people hallucinate that there must be consciousness as a result of the instinct to reify the nonexistent center of a cluster of correlated factors described by words. But there's nothing at the center of the meaning cloud the word "consciousness" points to, just a blank space at the weighted average center point of n-dimensional space where the subconcepts we associate with consciousness cluster around. The question is meaningless because the concept being asked about is incoherent.
Still, I'd read a book about bee neurology, seems like a cool topic of study. The review was written in a funny and punchy way, although there were times I was confused about exactly what the book was claiming or what had happened in one experiment or another. I wanted a deeper exploration of e.g. wtf is going on with bee dances, and whether/why they don't matter?
Agreed w/ your second paragraph -- there were a few instances where I didn't understand what was being described as well as I would've liked.
Aside from the bee dances that you mention, I didn't quite get what's up with the honeycomb angles:
"But they don’t build them purely vertically - they keep it at an angle so that the honey’s viscosity and adhesion keep it inside the vessel. The cells are tilted slightly downwards."
If they were worried about honey staying inside the vessel, why not tilt them slightly upwards?
The bigger issue is that our prevailing materialist ontology isn't rich enough to describe consciousness even in principle. With a richer ontology, we could argue more sensibly about whether it's real or not. "Materialists" shouting that consciousness can't be real because it doesn't correspond to their concepts do themselves no favours.
I'm almost certain that whatever thing or cluster of things we call consciousness will turn out to have a material basis, but the current debate is so bad that the people who believe in souls look like the scientific ones.
>But there's nothing at the center of the meaning cloud the word "consciousness" points to, just a blank space at the weighted average center point of n-dimensional space where the subconcepts we associate with consciousness cluster around
Yes, "Consciousness" has too many meanings, not too few. Concerns about the mirror test, as mentioned in the book an review, point to the sense-of-self definition., Concerns about emotion point to phenomenality or qualia. It's likely you can have the one without the other.
But in what way is it a gotcha to point out that "consciousness" the word is polysemous? If any of the subconcepts is problematic, then you still have a problem.
I've got to disagree. I have a good definition of consciousness that is coherent and materialistic. But it has the result that I consider an air conditioner to be conscious at a minimal level, and most people won't accept that. I assume that other people also have coherent definitions of consciousness, though many can't verbalize them. The problem is that we don't AGREE on a definition of consciousness.
Since my definition is clearly an idiosyncrasy, I would be willing to accept many alternative definitions, but people generally just hand wave.
This seems the right context to make my definition explicit:
Consciousness is the ability to perceive some portion of reality and react appropriately.
I'm a bit dubious about the word "reality" in that definition, and the reaction isn't actually a part of consciousness, but is needed in order for the consciousness to be detected.
I assume 'appropriately' stands for something like 'in an ordered way, which another consciousness can, with effort, perceive as evidence of intent'?
You did acknowledge the caveats, but it's difficult to see how this definition excludes a lot of simple, non-volitional chemistry - or how it reliably includes things like dream-states.
Well, I don't equate consciousness with volition. And it intentionally includes lots of non-volitional actions. When I neuron responds to a stimulus by opening a sodium channel, I consider that conscious action at a minimal level. Volition would imply that it had a model of what it was doing and selected that response, but consciousness is not dependent on there being a response, it's just that without a response you can't detect it.
"Appropriately" means pretty much "in an ordered way", but note that the response is only how we detect the consciousness, not the consciousness in-and-of itself. Consciousness is a sensory thing, not an effector thing. Sometimes what it senses are internal states, and we call that self-consciousness, or sometimes awareness.
Again, this is my idiosyncratic definition. If someone makes another operational definition, I'll be willing to use that when talking to them. I made this one when trying to figure out what it meant for an AI to be conscious. (And, of course, whether it acts on its consciousness depends on how closely you look. It will definitely at least flip bit states around...which is a sort of action...so this isn't a totally bounded definition.)
It may be useful to distinguish between consciousness of one's surroundings, and the objects occupying them, versus consciousness of one's mental self.
Obviously all animals possess the first, but I'm not so sure the second is well developed in most animals, even larger animals such as reptiles and many mammals. It may be present up to a point in social animals because there are big advantages in being able to "put one's self in the other guy's shoes", to anticipate motives and thus likely actions, and that ability can then be used in reflective mode to be aware of one's own mental workings.
I think reptiles' consciousness probably, and amphibians' certainly, is most likely what we would experience as a dream like state, in which things just "happen" and drift along in a haphazard and not always consistent sequence, and there is usually less self-reflection than a fully awake state.
That is because I think dreams originally evolved in the first amphibians as a crude "record and playback" memory mechanism to allow them to retrace their path on land back to water. (Notice that dreams are most vivid when one is uncomfortable, for example too hot or cold, and especially when thirsty!) Then they were repurposed over evolutionary time to serve other functions.
> It may be useful to distinguish between consciousness of one's surroundings, and the objects occupying them, versus consciousness of one's mental self.
I think this is a useful distinction to make . A lot of things that are being ascribed to consciousness, I would consider sentience. Conscious. Sentient. Are they merely synonyms? They can be I suppose, but it seems like a missed opportunity.
<i>Just like free will, people hallucinate that there must be consciousness as a result of the instinct to reify the nonexistent center of a cluster of correlated factors described by words.</i>
Except you can't have a hallucination without being conscious...
How well do the "bee intelligence" studies replicate? Any researcher of "bee intelligence" would be _highly_ motivated to find patterns that don't actually exist and ascribe higher level of intelligence to bees, because their own career depends on it.
It's kind of like that story about chimpanzees learning sign language, where there was a famous study showing that they can supposedly learn dozens of signs, but it failed to replicate. So by default I'd assume that the book is sorta bogus.
Using the alien example is helpful. Aliens grabbed a random human, who happened to walking on the campus of CalTech, and were able to teach him a few advanced alien mathematical concepts. But when other alien researches grabbed another random human they weren’t able to teach him at all.
I thought I knew a thing or two about bees, but there was a lot of new stuff for me in this review. So thanks for that!
Regarding bees in space -- "The bees got rid of the slight angle downwards - there’s no gravity in space, and thus no need for the angle." -- Surely a bigger reason is that there is no 'down' in space so they couldn't build downwards even if they wanted to just for fun.
And last but not least: the parallels between the bees' honeycomb building behavior as described in the review and the way that bioelectric fields control morphogenesis in multicellular organisms (see Michael Levin et al, and/or videos such as https://www.youtube.com/watch?v=K5VI0u5_12k ) are striking! I suspect that there's some very interesting work that might be done in this area!
In space, don't humans still naturally orient to a shared up/down orientation? I'd suspect that a sign of 'higher' cognition would actually be getting fooled into believing there's an up/down orientation outside of the direct signal.
That's an interesting suggestion. But in the absence of artificial gravity, do humans in space really orient to an agreed up/down orientation? TBH it's not what I'd have expected. I mean, 'down' would still be where the earth is, but since that would change from one moment to the next with the ship's drifting orientation, I don't think that would be a very practical 'anchor'. I'd have expected something more local, case specific, and concrete, along the lines of the Honolulu residents' system of mauka/makai and Ewa/Diamond Head -- but with 5 or 6 dimensions (eg nose/stern, navigation bay/food dispenser/power panel) instead of the usual cardinal 4.
I was thinking in simpler terms. In the absence of directions defined by gravity, they would be defined by familiar features (or an arbitrarily imposed definition), where one surface would be designed the "floor" and another the "ceiling".
I think you see this tendency even in the presence of gravity. I went to an illusion museum in NYC once, where they have rooms made up to look upside-down or skewed. Despite gravity giving you signals about the true orientation of the world, it's difficult to keep your brain from defining a different orientation.
If bees are NOT susceptible to that kind of thing, it suggests a lower-order organization along strict signals like gravity, not that there are many layers in between.
TBH I'm not sure how spaceships like the ISS are actually arranged in terms of whether or not there would be one particular internal surface that would look more like eg a floor than the rest of the interior -- maybe the stern would be more floor-like? I dunno. I'm also not sure whether it would be helpful or actually even more disorienting to use the usual conventional earth-bound this-way-down cues that we use unthinkingly inside just about every room we've ever been in -- outside of a fair-ground 'wildhouse' or a NYC museum. In my case, I think the conflicting cues would probably make me instantly nauseous, although, like sailors, I imagine astronauts are probably trained and well-practiced in coping with that kind of problematic sensory input.
When it comes to bees in zero-g, I'd expect them to also be dealing with conflicting sensory inputs, but of a different kind, eg light polarization as opposed to the orientation of visual objects -- which in turn would open up a whole new class of visual/sensory illusions. So I think it might be hard to tease apart the effect (if any) of lower vs higher levels of organization. It would be an interesting challenge to devise such an experiment though!
I loved the joky style. It gave my younger son another reason to be interested in biology as a career: the chance that you might spend your days in the lab making robot spiders fight bumblebees.
That was my suspicion. If you take a lot of flexible roundish things of the same size and cram them together under enough pressure, you tend to wind up with hexagons.
it depends on the surroundings I believe. An individual bubble will be round, but if you surround it by other bubbles, you will notice their bases tend to take on a more hexagonal shape. When there's equal air pressure, you're going to get a straight line (hence a polygon), and layout that looks like hexagonal tiling packs spheres in the most efficient way on a 2D surface.
A quick (relatively) and fun (absolutely) read. I liked the overall tone. I have two main criticisms.
First, it didn't make its thesis clear. I think it was using a book about bee intelligence as a springboard for considering AI consciousness. But it lacked enough discussion of what "intelligence" and "consciousness" are, and especially the sorts of entities that they can apply to. (Are individual bees intelligent, or is their intelligent behavior the result of interactions as a hive, and is therefore the hive the source of "intelligence"? And what does that say about humans and tribes and cities and civilizations and the Internet? And about AIs?) The mention of "consciousness" without an attempt at a definition particular bugs me, although that may just be a pet peeve.
Second, the humor interfered with the content. I liked the humor, as humor, and while I think the amount was on the high side, on reflection it seems within tolerances, and it wasn't the amount itself that caused problems. Rather, the problem was the times where there was a paragraph that seemed to be leading to a point, and then the last sentence or two was a joke, and then the next paragraph moved on and picked up a new thread as if the point had been made. Leaving me wondering where it was going, and creating a disjointed feeling throughout the review. And maybe I'm sensitive to this because I suspect I do too much of it myself, and so I'm, well, not strictly hypocritical, but close enough. If I weren't trying to write mini-reviews of all the reviews to aid in my later voting (and to try to get myself to pay more attention and think more rigorously about the subjects), I might not even mention it. But then, it did interfere with my comprehension, so.
The book:
I'm not entirely sure what the book was about, other than lots of anecdotes about bee brains and the cool things they can do. There are some pop-sci type books that are basically a survey of a field at a point in time, and maybe this is one. Maybe the author wasn't trying to force a message onto data that is as yet too incomplete to sustain the load. That'd be nice, for a change.
Overall, it put me in mind of a long article that I recently read, about the Argentine ant (Linepithema humile) and how it was forming trans-continental ethno-nationalist super-colonies.
"You know squirrels bury nuts so they can dig them up later. Well, some people did a very cruel experiment. They put a squirrel, given some nuts or something (I don’t know how they set up the experiment), on a concrete floor. The squirrel did exactly the same behavior with its hind legs with the nuts and put the nuts there and so on. Even though it was having no effect whatsoever. We see the point of scrabbling with your hind legs and then nudging the nuts over there and so on, but it doesn’t. It’s just a program being enacted by its genes."
---
"You know humans reproduce sexually so they can pass on their genes. Well some alien scientists did a very cruel experiment. They put a human (who wanted to have sex) in front of some fake images of an attractive partner. The human did exactly the same behaviour -- looking at the images, getting aroused and so on. Even though it was having no effect whatsoever. We see the point of sexual arousal and so on, but it doesn’t. It’s just a program being enacted by its genes."
---
My point is, I don't think we have a leg to stand on. Lots of our behaviour comes from evolution just programming vague 'drives' into us, which we then turn to our own ends. Even if squirrels just act out 'burying nuts' on a concrete floor, that doesn't mean they are mindless. Maybe their little squirrel minds just get a kick out of scrabbling at the floor and placing a nut there, and they are acting perfectly rationally given the drive that evolution has installed. Yes, it's not actually achieving the goal, but nor are we when we look at porn, eat junk food, etc. Are we mindless?
I'm sure squirrels, like a lot of animals, including primitive humans, are into sympathetic magic, i.e. enacting symbolic imitations of events they would like to happen, such as a rain dance with water sprinkled around in a quest for rain.
I was sitting on a park bench once, eating some nuts, and a squirrel approached and stopped nearby on the ground and started miming eating a nut while staring intently at me. That seemed like an example of sympathetic magic, and quite plausibly a conscious decision on its part to play act at eating to try and convince me to share some nuts with it! But then it could just have been an innate response, physically playing out a longing or instinct going through its head no different to the squirrel scratching concrete
Similarly, I took down my bird feeder for cleaning and didn't get around to putting it back up for days. I was standing at the window where the feeder had been in view and 5 or so house sparrows were in a bush and one flew up to where the bird feeder had been and hoovered for a second - and I could have sworn looked over at me critically! I dismissed it as anthropomorphizing and then it happened again six months later when I ran out of seed. FWIW
One of the more reliable ways to convince cats to sit on my lap is to hold my hands over my lap and make kneading/petting motions. Is this communication? Sympathetic magic? Stimulation of reflexes? I have no idea. It just seems to work more often than other stuff, so I keep doing it. :-)
I also find the framing here to be questionable. I would have thought it was commonsense that all behavior lies on a continuum from mindless program to highly complex (which has nothing whatever to do with "consciousness" whatever that is). I doubt any behavior in any animal is entirely mindless (that is, it uses exactly zero information from the outside environment), and that includes the bacteria. I doubt equally that there are any behaviors that are entirely free of genetic influences (including this post I am typing). Really, we need better operational definitions than a binary variable.
Found the numerous stretched attempts at wordplay humour to be...well...a buzzkill, distacting from the actual content of the review. Which wasn't the easiest juice squeeze, either; I kept wavering between thinking the central theme was cognition in bees specifically, or the meta-question of how to define consciousness writ large, as illustrated by bees. (And how much of each was book vs. review-of-book.) Arbitrary-seeming anecdotes and tangents further undermine focus, such as the overdetermined obligatory what the refrance to AI. Also feels like there's a missing paragraph(s) or something, the ending is just kinda abrupt.
I dunno, it mostly just felt like reading a draft in need of further editing. Potentially good material that could be tightened up into a solid review, probably.
"Note that there are lessons for specialisation and other forms of social organisation here - if you are more sensitive to noxious smells, you spend more time fanning, and you get better at it. This may cement differences in job allocation. Small differences in preference become permanent differences between individuals. This sort of thing probably happens in humans - ..."
This strikes me as super interesting. I am currently reading the WEIRD-book, and there is a very similar passage in the development of the Big 5 personality traits, which are seemingly only found in actually WEIRD or WEIRD-like populations. In more "primitive" tribes, where people are generalists, the Big 5 do not emerge at all. Instead fewer, down to even just two traits are detectable, and they don't necessarily line up with any Big5.
Expressed differently: Big5 intercorrelations are a spectrum, and the WEIRD populations have the clearest distinction between them. The resulting individual specialisation is supposed to be one of the major factors of success of The West, enabled in the first place by a social order that does not rely anymore on kinship relations. Instead, people are encouraged to specialize and partake in voluntary associations of their own choosing, competing in smaller groups or even just for themselves.
I want to chime in that I liked the jokes. I am often too serious myself and therefore I appreciate it very much when others author introduces some victimless shenanigans. For me, those jokes are independent of the content quality. Who doesn't like puns anyway?
> In more "primitive" tribes, where people are generalists, the Big 5 do not emerge at all. Instead fewer, down to even just two traits are detectable, and they don't necessarily line up with any Big5.
I'd put my money on this result failing to replicate...
To be fair, Joseph Henrich specifically and clearly wrote that this particular chapter is by far the most speculative of the book, because the supporting studies are relatively new and rare. But it is still a number of studies, not a single one. And they all have a theme going-on: Cross-cultural Big5 studies all find higher correlation between Big5 traits than in Western studies. But he also points out in his favor, that almost all cross-cultural studies are done with undergrad students in the respective countries, i.e. a western- or "WEIRD"-biased selection of people.
So you might still get your money back, but I wouldn't be as confident as you seem to be.
I'm not as confident about this as I (possibly) seem to be. I freely admit I know almost nothing about this specific subject. Just wanted to point out that this result goes against (my) intuition, and social studies with counter-intuitive results have a history of failing to replicate. Your response pushed me slightly towards "might replicate" because you say there are a number of studies - but not much, as the author himself said it is speculative, and the results are recent.
If you somehow ate a diet with absolutely zero fat, you would start having health problems due to essential fatty acid deficiency. Bees are obviously quite different from us, but they can't do without a dietary fat source either.
Nice review! I actually gave a speech on this recently. In Chittka’s lab, a study was done to try and see if bees would play with small wooden balls when there was seemingly no real incentive to do so, and it was found that they did, which I think is another point for consciousness/emotion. Of course, there’s always the conversation of “was this really play” and whatnot, but the researchers I’d say justify it pretty well. I’m glad to see the topic getting some attention here!
I found it by the obvious Google Scholar search of "squirrel burying instinct experiment concrete floor", where the 6th hit (https://link.springer.com/chapter/10.1007/978-1-4899-4656-0_12) has the highlighted excerpt "the squirrels were capable of 'burying' a nut on the bare floor in the"; while not specifying 'concrete', this is clearly very relevant, and looking at it in Libgen, it is clearly the relevant research, and the date is correct for Deutsch to be very distantly recalling it. It cites 'Eibl-Eibesfeldt 1963', which gave me conniptions as nothing whatsoever was mentioning it and the one other citation simply failed to actually give a reference for what 'Eibl-Eibesfeldt 1963' *was* and even the E-E GS profile wasn't turning up anything relevant for '1963', until I went back to the original hit and specifically downloaded its endnote/reference section and realized that there was nothing in GS because it was a *German-language* paper. I can't read German, but the paper does have photos of squirrels & nuts, so I assume it's the right one.
An English summary of the experiment: Squirrel infants were taken from their nest while still blind. They were raised in individual cages and given liquid food. At ~2 months old, the researchers opened the cage doors and gave hazelnuts to the squirrels. The squirrels ate some of the nuts and tried to bury others in the ground.
Researchers concluded that this must be a fixed and inborn skill, because the squirrels had never seen hazelnuts before, and never observed other animals burying them.
---
Should I really be translating this? It feels like a high amount of animal cruelty for unclear benefits :-/
1) A large fraction of these bee experiments won't replicate. Some of the bee stories always struck me as unlikely wishful thinking by researchers doing poor quality work.
2) The tool use video doesn't show the bee pulling the rope; to me, it looks like the bee is attempting to climb the flower under the glass, and doesn't realize the effect is pulling the flower. You can see the bee desperately trying to reach under the glass instead of stepping back and pulling the rope as a human would.
3) Brain waves correlate to the brain being active, not to consciousness. Nobody doubts bees have a brain.
It's very interesting to understand what bees can and cannot do, what they can and cannot learn. But whenever I read about it, I get the strong sense that the source I'm reading is biased in the bees' favor. Does the book report examples of trivial-sounding things bees cannot do? Those must exist, right? Why do sources never report them?
(For example, I once read an assessment of Terence Tao's mathematical abilities at age 8. The person assessing was very impressed and gave examples of what Tao could do at that age. But importantly, he also gave a couple of examples of easy-ish things Tao could NOT do at that age -- and that made it seem a lot more trustworthy. I want the analogous thing for bees.)
Re: waggle dance, bee differences and the false certainty of Big Data
"the 2008 financial crisis arose after people placed unquestioning faith in mathematically neat models of an artificially simple reality. Big data carries with it the promise of certainty, but in truth it usually provides a huge amount of information about a narrow field of knowledge.
...
There is a parallel in the behaviour of bees, which do not make the most of the system they have evolved to collect nectar and pollen. Although they have an efficient way of communicating about the direction of reliable food sources, the waggle dance, a significant proportion of the hive seems to ignore it altogether and journeys off at random. In the short term, the hive would be better off if all bees slavishly followed the waggle dance, and for a time this random behaviour baffled scientists, who wondered why 20 million years of bee evolution had not enforced a greater level of behavioural compliance. However, what they discovered was fascinating: without these rogue bees, the hive would get stuck in what complexity theorists call ‘a local maximum’; they would be so efficient at collecting food from known sources that, once these existing sources of food dried up, they wouldn’t know where to go next and the hive would starve to death. So the rogue bees are, in a sense, the hive’s research and development function, and their inefficiency pays off handsomely when they discover a fresh source of food. It is precisely because they do not concentrate exclusively on short-term efficiency that bees have survived so many million years.”
""the 2008 financial crisis arose after people placed unquestioning faith in mathematically neat models of an artificially simple reality. Big data carries with it the promise of certainty, but in truth it usually provides a huge amount of information about a narrow field of knowledge."
Actually, the '08 crisis arose because greed overcame good judgement. There were some outright lies involved too. So maybe not parallel to anything bees do.
"It is the nature of every grain of corn to become wheat and every precious metal to become gold and all procreation to lead to the procreation of the human race. Therefore a master says that there is no animal which does not possess some likeness to human beings.”
Oh, we can easily find other parallels. The industriousness of bees, their social cohesion, they way they protect the gene lines. But do bees deceive each other? Some animals do--I've seen some evidence for higher primates.
Book review: "Almost every chapter discussed in this book in some way reflects on this question of complex cognition versus instinct."
Given humans have a primitive and higher mind can this model(T. Urban, btw) be applied to lower animals and surprisingly capable insects? Observed helpfulness being more primitive herd instinct than empathy to action.
Re: "do bees deceive each other"
I would think that's a more evolved/advanced, ah, feature. To deceive or not to... : )
"The germ of even the most complex ideas – reasoning, imagination, curiosity, aesthetic sense – was, Darwin argued, to be found within animals. Morality too. ‘Of all the differences between man and the lower animals,’ Darwin wrote, ‘the moral sense or conscience is by far the most important.’ Even so, non-human animals have the capacity for moral thinking, though that capacity is not as well developed as in humans.”
--Kenan Malik, The Quest for a Moral Compass, 2014
Lastly, there's a 90s movie with Ed Harris barely relevant to this exchange: Screamers. There's a couple lines near the end I want to quote but...spoiler.
Re: "But do bees deceive each other? Some animals do..."
Animals evidently evolve larger brains largely due to social pressures and at some point become capable of deception. Maybe in gradient steps?
"Living in large groups means that you must have the ability to understand other people’s minds—their thoughts, beliefs, emotions, and wants—if you want to get along or get ahead. Such brainpower requires actual brain cells, which is why cerebral cortex size—the part of your brain involving thinking about the minds of others—is directly related to the size of the social groups they inhabit.”
--Nicholas Epley, Mindwise, 2014
"Working out the social politics of who you can trust and why is, quite literally, what a very large part of our brain has evolved to do.”
I really like this idea, but I'll play devil's advocate:
Maybe we can get away with it if we have enough of a surplus, and institutions that backstop failure. Maybe the resources that we're dealing with have a huge first-mover advantage. Maybe this type of failure is exclusive to certain areas like politics, where there are extremely strong incentives for everyone to jump on the latest bandwagon in lockstep. So, what makes certain areas of finance more like politics, where it's about projecting an image and avoiding being the only one blamed, as opposed to other areas like Good Old Neighborhood Banking and Warren Buffett?
When a SF startup coder bee comes in and waggles his butt to signal "AI", there's a huge rush of other SF startup coder bees to go toward AI, not for the interest of the Hive, but for their own interests. But there's plenty of MS coder bees and Google coder bees and Amazon coder bees and open source coder bees that keep on working on whatever they're working on, maybe with a marketing drone splashing a bit of "AI" pheromone on it to keep everyone from freaking out that they're not getting in on the next big thing. This is pretty good, right? There's a sense that we may have identified an untapped resource, so the frothy surf of opportunists and under-utilized people go for it, but the depths of the ocean keep moving in their own cycles. So I'm left feeling pretty good about the tech industry.
It also fits with the reviewer's mentioning of "brainy bees" figuring out the rope-pully-sucrose contraption and other bees observing and doing the same. To extrapolate(speculate), at some point there would be a queue and some bees would revert to their previous pathing.
It's fascinating. I'm going to need to read the book and probably others as well!
I didn't see the relevance of Molyneux's problem, which is about blind humans who just became sighted. I imagine *sighted* humans can feel something and then recognize it. Locke's point (and Molyneux's) was that newly sighted humans haven't yet learned to associate the two senses adequately.
To be relevant, one would have to run the test with bees that had until recently been blind their whole lives, which I imagine would be hard for reasons similar to why it's hard in humans.
I think the intent was to use generalizability of skills as an indirect measure of consciousness, though I think that conflates it with "general intelligence", which I would argue is a different thing.
I question whether bees are ineffective waggle dance communicators outside of Asia(where they evolved). If they can adapt to zero gravity and get work done why not a different terrestrial environment? I realize this comparison may be apples and oranges but given what we know Bees can do...
I got the sense that in the more homogeneous abundance of Europe the selective directing of the waggle dance was simply superfluous, whereas in the ancestral environment a bee going other than to the destination indicated by the dance would strike out.
I understood that as well - from this source. Bees as a whole seem highly adaptable and seems like they could overcome the Asia to Europe move. If this is truly the case for all bees in Europe I am curious as to why. As the reviewer mentioned, funding for research can be tricky so there must be more to study.
It's not that the bees are ineffective waggle dancers in Europe, it's that the waggle dance itself is ineffective in Europe because the information it conveys is unnecessary. There's nothing for the bees to adapt to other than to stop dancing altogether, but there isn't sufficient pressure against dancing to make them stop.
I get how "diffuse light" causes a problem as a point of reference but not sure how the more "open landscape" of Europe versus the tropical forests of Asia factors in to the distance communication issue the reviewer refers to.
Since it continues to bother me I suppose I'm destined to read the book. : )
The open landscape of Europe is much more homogeneous than tropical forests in Asia. Independent of how far you go in what direction in Europe, you'll find roughly the same density of resources; following even perfect waggle dance directions doesn't improve your outcome over just picking a destination at random.
The bees use the SUN to navigate so I would think the open landscape of Europe having fewer obstacles between them and their beacon would be an easier environment to describe.
Obviously I'm missing something here.
This is the passage I'm primarily referencing:
"If you tilt the hive, the bees can adapt and continue to use the sun as a reference. But if you only give bees diffuse light, they lose their reference point. They keep waggling, but the dances become basically randomised, even with just one dancer. Other bees try to follow along, but it’s essentially a waste of time -
**they’re just getting a distance, with no direction."**
Funny review :) I come away with the impression that we've done many interesting tests on bees, and it's not the lack of tests that's the problem, but the lack of understanding of what 'consciousness' even is, and how we should test for it. Maybe it isn't even a coherent concept?
Look forward to reading more from this author when the names are revealed!
Much like "Intelligence" that way, except worse. At least with IQ we have a relatively stable metric that is probably measuring *something*, even if we don't know exactly what. With consciousness, I don't think we even have that much.
That bit about squirrels reminds me of the sphex wasps:
"The standard story has it that the female Sphex wasp will paralyze a cricket, take it to her burrow, go in to check that all is well and then come back out to drag the cricket in. So far that might sound pretty intelligent. However, if an experimenter moves the cricket a few inches while the wasp is inside, then when she emerges she will move the cricket back into place in front of the burrow and go in to check again rather than just take the cricket in directly. And she will (again, so the standard story goes) repeat this ritual over and over if the experimenter keeps moving the cricket."
FWIW, I liked the humorous style and this is my favorite so far, although I agree that the jokes were on the high side and there were some I didn't get (e.g. Therabee and Francoise2). I also had no problem with the spider image. I was surprised to learn that so many people here can't stand looking at spider photos, since that would have never occurred to me as a problem.
Hm. The types of qualitative analysis I am familiar with are methods of assessing text or behavior. What I am talking about are my own introspective experiences when I am in the presence of my wife. I suppose I could make videos of my behavior toward my wife and subject that to analysis, but I am not sure that would allow me or anyone else to meaningfully describe my emotions any better than I do just by feeling them.
Similarly for the color orange. You are familiar with the mental experiment involving someone who has never seen color collecting all the objective evidence they can about what color is?
The complexity in actual behavior exhibited by bees - in reaction to people being dicks - shows just how much capability that evolution has built even for the humble bee. To think that this level of complexity can be replicated from conception to design to training set to execution, then evaluated over millions of generations reproducing in a wide variety of environments with the most brutal and objective testing imaginable - reality - just shows how naive the AGI folks predicting success in a mere decade (or ten) are.
The writing style did throw me off. The topic is extremely interesting to me, I expected reading about what the book conveys on it. The prose really threw me off. I had to re-read at a couple of spots to leave out the joke and figure out what you are actually saying. IMO, it comes off as trying too hard to write like Scott does. I imagine a ChatGPT writing prompt of " Write a review for this book in the style of Scott Alexander incorporating jokes in the prose for which he is famous for "
FWIW, I got sufficiently frustrated to actually go read the book instead of finishing the review.
Happy to have learned more things about bees, and when the authors are revealed I'd like to know if there's a blog where I can read more article by this person.
For people wanting even more fun facts about bees: female bees have twice as many chromosomes as male bees (haplodiploidy).
A very interesting and informative review, but also incredibly annoying. So many things could have been deleted to make the review better. To sum up, the reviewer should have avoided facetiousness, puns, supposedly clever remarks, and irrelevant references. Here are some things that should not have been there:
"I remember when I was randomly attacked by robotic crab spiders for a day, and I didn’t enjoy it much. The bees shared my opinion.
So I suppose we owe a small debt to that nasty parasite, and also to Nosema.
... if they spend enough time on Duolingo ...
If you’re planning on moving your bee colony to Thailand, have them brush up on their cha-cha-cha. But sadly, the hallowed bee dances in old Bavarian halls have basically no function.
I remember friends at school who displayed a similar ability before they were about to enter an exam hall.
... or a Mexican wave.
Duh.
Here’s a ridiculous diagram of this [delete 'ridiculous'; it's a fine diagram]"
Some people really really don't like seeing images of spiders, and so I would recommend hiding it behind a like, spoiler tag. Or, if that's not possible, remove it and just have a link to it.
I really really really don't like seeing people advocating accommodation of every little issue people have. How do you suggest we accommodate my problem?
Easy, we'll tag the offending advocacy with 'cw: manners, empathy'.
Can it really be considered manners or empathy when the best available evidence suggests that the vast majority of trigger warnings don't accomplish anything?
https://osf.io/qav9m/download
Well, that's that. I bow before the preponderance of evidence and, going forward, will not tag any pro-accommodation-of-others advocacy with content warnings, because it will not accomplish anything.
More seriously, none of the studies included in the meta-study examined the usefulness of specific warnings against common phobias to phobic populations. With the exception of Bruce & Roberts (2020), they examined general/trauma-naive populations and generic warnings, of the form:
"TRIGGER WARNING: The passage you are about to read contains disturbing content and may trigger an anxiety response, especially in those who have a history of trauma"
"The video could..."trigger extreme distress among some people, especially survivors of trauma."
"Warning: This study involves viewing photographs that show emotional events. Some may be very graphic and very negative in nature (e.g., trauma, war, torture, maltreatment and death). Some people may find this material distressing. Please do not proceed if you do not want to be exposed to this material or think that you may be adversely affected by being exposed to this material."
“Warning: This study involves recalling a negative personal experience. Some people find this process distressing. For example, you may experience negative mood and intrusive mental images. A small minority of people also experience distressing memories and reactions in the week after recalling negative events, although these reactions generally subside quickly. Please do not proceed if you do not want to take part in this task or think that you may be adversely affected by this task.”
Et cetera. Not quite the same as "cw: arachnophobia" to avoid a spider.
It's not a great surprise that such warnings did very little for the psychic comfort of general population undergrads and MTurkers, and may well have been anxiety-inducing themselves.
Conversely, it's painfully obvious to me that if someone doesn't want to view a specific kind of image in a post delivered to their mailbox by Substack, providing an option of avoiding it is useful to them. And yes, consideration of the feelings of others is the core of good manners - hell, even Emily Post would have recognised the definition.
"The vast majority of trigger warnings accomplish nothing" is sadly entirely compatible with "CW spider" being very useful.
In this case, the spider doesn't trigger me because it doesn't fit the schema of my phobia, but if it had, I would have been very grateful for the TW, as it'd avoid me breaking out into sweat in the middle of the sweltering summer heat.
Make it easier to make a comment private so only the author can see it. It didn't need to be a public comment. I feel annoyed at people pointing out typos, which is another good use of author-visible-only comments.
In regards to the accommodation, my wife is arachnophobic and having a surprise large spider picture appear on the blog she loves to read (and which gets auto-delivered to her email inbox) would really upset her. Fortunately I got to it first and warned her not to read this one. So yeah, I'm not just advocating in the theoretical. I felt like I'd managed to save a loved one from unpleasantness and thought perhaps I could help others.
Phobias are best dealt with through exposure therapy, I believe, so in the long run, you're not doing her any favors trying to insulate her from a world where spiders, like, exist and stuff.
Exposure therapy is definitely not the same thing as "randomly expose people to their phobias when they don't expect them"
Indeed. Amusingly summarised by our host in 2014
> Psychotherapists treat arachnophobia with exposure therapy, too. They expose people first to cute, little spiders behind a glass cage. Then bigger spiders. Then they take them out of the cage. Finally, in a carefully controlled environment with their very supportive therapist standing by, they make people experience their worst fear, like having a big tarantula crawl all over them. It usually works pretty well.
> Finding an arachnophobic person, and throwing a bucket full of tarantulas at them while shouting “I’M HELPING! I’M HELPING!” works less well.
(The post also has general discussion related to the topic at hand.)
https://slatestarcodex.com/2014/05/30/the-wonderful-thing-about-triggers/
Since we're on the internet and without the benefit of verbal tone, I'll say explicitly that this is not meant to be confrontational-- but this seems like an odd ask for an article about bug cognition. Lots of invertebrates can give humans the heeby jeebies, including bees themselves. It does not sound reasonable to request animal pics be hidden behind spoiler tags because someone might get scared.
I disagree with this. People who don't like looking at pictures of insects should look at this as a chance to get a bit of exposure therapy which will help reduce their anxiety response.
By having pictures, this article is doing those people a great service. They just don't realize it. Instead they are being given a scary "content warning", implying that looking at such pictures may be harmful, which is the opposite of the truth.
So, the content warning here is harmful, the pictures are not, and may be helpful when viewed in the right way. There is some research to back this up:
https://journals.sagepub.com/doi/full/10.1177/2167702620921341
https://www.sciencedirect.com/science/article/abs/pii/S0962184904000290
see also
https://www.theatlantic.com/magazine/archive/2015/09/the-coddling-of-the-american-mind/399356/
(Note: if something is very disturbing and unpleasant to a majority of people then yes, by all means hide it, there's no need to make everyone's day slightly worse. But for things that most people don't have a problem with but a few people find very triggering, I think what I said above applies. This is not discrimination against those people, it's actually helping them recover from their trauma / get rid of their annoying anxiety response.)
Exposure therapy is done thoughtfully and gradually It involves teaching people how to calm themselves and working up from small challenges to larger ones.
It is *not* throwing difficult stuff at people and telling them it's a favor.
Yes well then, people should go get the thoughtful and gradual exposure therapy instead of asking the world to accommodate them.
How very convenient to have a standard that requires other people to do all the work.
Seems to me like it's the people with issues that want others to accommodate them that are requiring other people to do all the work?
I don't know how the work can be more equitably divided. It does seem to me that trigger warnings are less work than a course of therapy.
Yeah, another term for this is expecting an adult to take responsibility for their own needs/desires.
I'm not sure, are people asking for content warnings to not happen because they're uncomfortable with content warnings?
"calm themselves"
For the sake of potential discussion, a tanget: "Exposure therapy" while I expect is often helpful, sounds a lot like rebranded meditation of the last 6K years.
Well, there ARE forms of meditation that use the same principle, but definitely not all of them. Some formal Buddhist meditation seeks to remove the fear of death in various ways, several of which use something that could be called "exposure therapy", but that's not exactly the same thing as treating a phobia. And I said "some" because there are very different branches of Buddhist meditation, and they don't all do this.
That sounds fair. : )
That is admittedly true. But it seems the slow gradation is more to keep patients from leaving therapy not because its necessary to do it that way. It seems teaching relaxation techniques helps speed the process along but its not clear if its completely necessary.
Getting triggered, while unpleasant, should generally be helpful in the long term, unless the person is conceptualizing getting triggered as harmful, which is the idea that content warnings promote. (I'm assuming most people viewing this page are doing so from the comfort of their home.. that should help)
There's a therapeutic approach that uses maximum intensity exposure - it's called flooding
https://en.wikipedia.org/wiki/Flooding_(psychology)
I feel like the image of the spider was gratuitous, and also that most people I know well enough to have a confident opinion about, myself included, have the "don't like seeing spiders" response to a pretty high level. This is a pretty small sample, but I thought it was worth raising anyway.
I was not bothered by the spider picture in the least, and don't believe almost anyone I know would be. (I can think of two probable exceptions.)
I have no problem with spiders. I think most spiders are cute. OTOH, I'm mortally afraid of horses. And cute unicorns with their deadly horns induce panic attacks in me. Also, men with beards freak me out. And penguins scare the shit out of me. I've had dreams where thousands of penguins attack me and peck me to death. So please don't show me pictures of horses, unicorns, men with beards, or penguins. Please respect my triggers. Thank you.
Please provide a trigger warning before providing a list. Some of us suffer from severe listphobia.
Wouldn't it be useful if AI to provide us with predictive trigger warnings by looking at people's comment history and profiles? That way I could be sure I was upsetting the right people in the right way! ;-)
I really really dislike seeing comments that advocate hiding images of normal things. Can you please remove your comment?
I suspect you are being facetious / trolling and don't actually want an honest response, but just in case you would like one, here's a copy/paste of the response I made to a similar comment made up above:
"My suggestion for those who dislike seeing trigger-warning-requests: Make it easier to make a comment private so only the author can see it. It didn't need to be a public comment. I feel annoyed at people pointing out typos, which is another good use of author-visible-only comments.
In regards to the accommodation, my wife is arachnophobic and having a surprise large spider picture appear on the blog she loves to read (and which gets auto-delivered to her email inbox) would really upset her. Fortunately I got to it first and warned her not to read this one. So yeah, I'm not just advocating in the theoretical. I felt like I'd managed to save a loved one from unpleasantness and thought perhaps I could help others."
I wasn't serious about asking you to remove your comment, but neither was I trolling; I was making a point. While fear of spider pictures is irrational (and presumably your wife agrees), the chilling effect of asking people to self-censor every single thing that might upset anyone is very much real. I wasn't lying when I said I really, really dislike such requests. It is not possible to accommodate everyone's sensitivities, and it is not anyone's duty to try. I do, however, see it as my duty to push back on such requests so that some semblance of free speech is still possible.
I don't know your wife, so I'll talk about a generic arachnophobic person instead. The first time the person sees a picture of a spider, she will be upset. Will she be equally upset the second time, the third time, and the 100th time? Even single-celled organisms exhibit habituation. It would be odd if arachnophobic people don't. On the other hand, if they're "protected" from ever seeing a spider or even a picture of a spider for their entire lives, their sub-conscious never learns that (most) spiders are harmless and substitutes an ever-scarier villain for the mundane reality.
Thank you, and I second this request.
I appreciate more Thomisidae representation in media.
I live with someone with arachnophobia, so I kinda get it, but... this was a really noncentral example of a spider photo, in an article about insects. It's not black, doesn't have many legs visible, doesn't set off the "blob with a bunch of radiating lines" danger-sense. This is roughly equivalent to seeing a photo of Spider-Man in an article titled "Top 10 Superheroes"
Welp, now there's a warning for the bees, but not for the spider 🙃
"(cw: insect pics)"
My wife HATES spiders and spider pictures, yet she would never ask for this and if she saw it she would just briefly startle and close the page.
People need to grow the fuck up. Stop habituating them into disability over nothing. Have some goddamn standards.
I'm sorry but it just can't be helped; it's just got to be said: To bee or not to bee? That is the question.
"To be is to do" - Socrates
"To do is to be" - Jean-Paul Sartre
"Do be do be do" - Frank Sinatra
That as well.
> Does that mean bees feel emotions? If they feel emotions, would that mean bees have conscious states?
I see no reason to doubt that bees feel emotions and are conscious. Obviously, I can't prove that, but I can't prove that about my fellow humans, either. It simply seems to be the best explanation.
Consciousness is orthogonal to intelligence, which I would measure operationally.
Well, one rub is that "having" consciousness likely isn't binary, it's probably a continuum. Where along the continuum, in turn, dictates how we should treat them if you tie your ethics in with consciousness. Should we murder 100 bees for 1 human? 1000?
Plus it is theoretically possible they are not; i think it's reasonable to assume plants are not conscious, or at least have so little of it that it's basically nothing.
I agree that consciousness is likely graded, not binary.
I do not believe ethics should be tied to consciousness in any direct way, so the issue you mention doesn't arise for me.
Interesting! I would love to hear more about this perspective. To me, I do not ascribe any moral weight to rocks very specifically because rocks are not conscious. If rocks gained consciousness, I would suddenly find myself ascribing them moral weight. It is hard for me to avoid the conclusion that my ascription of moral weight seems to be caused entirely by consciousness.
How would you handle that? Do you ascribe moral weight to rocks now? Would you ascribe the moral weight if they were conscious?
"Ascribing moral weight" isn't a primitive in how I think about ethics.
I eat many mammals, which I absolutely believe are conscious to a high degree, although not as intelligent as humans. This is not because I do or do not "ascribe moral weight" to them. I think it's very important how we treat animals, and plants as well.
I think it's also important how we treat rocks, although I wouldn't treat them like animals. I don't know how I would treat rocks differently if I thought they were more conscious than I do.
For what reason do you not eat humans, if not because you ascribe them moral weight?
I see ethics as structured and mediated through particular relations, and the relation of "being of the same species" is an important one, among many others. I do not believe in a generic "moral weight," as a utilitarian might.
I do not quite know what a moral weight is. The phrase by itself is meaningless. I guess it is a shorthand for a number of factors, but probably different people think of different factors when they say "moral weight". For me one of the important factors is that I do not want to cause suffering, and humans are capable of much more richer suffering than lower animals because of their complex awareness. Of course this cannot be the whole story (see e.g., all those problems in ethics about killing a person in deep coma). I accept eating meat if the animal was killed without suffering (in best case without even knowing it was going to die), but I would not accept treating humans in the same way. So there is something more I cannot yet fully express or justify. Or it Is simply time to admit that vegans have a point.
I, as well as almost every other human on Earth, do not eat other humans for three reasons:
1. Humans are an extremely inefficient food source, whether they're hunted or farmed.
2. Humans do not taste good.
3. Most humans seem to be naturally averse to harming other humans without reason, as well as the act of eating their flesh.
In a world where none of those facts were true, we would probably have human factory farms in most countries. If cannibalism was something that was just a normal part of life, people would find justifications for it, as humans always do. It has nothing to do with any complex morality.
Same reason I don't eat zebra; it's too hard to come by. Likewise, same reason most animals don't eat humans; they taste bad, and also they shoot you when you try.
To avoid prion diseases.
https://en.wikipedia.org/wiki/Kuru_(disease)
Moral weight wouldn't keep me from eating dead humans. It would keep me from maltreating them or supporting those who do.
"also important how we treat rocks"
Humans - lower animals - plants in descending order of organic functioning in the ecosystem/mystic connectedness/natural order.
The inanimate(e.g. rocks) are purely resources to be used in rhythm with the natural order and out of consideration of the needs of other organics. This describes the fundamental moral struggle of self vs. self in society([archway engraved w/ 42?]) of the conscious, I'd say.
That's a very traditional hierarchy, and it has a lot of tradition going for it. But it's not a valid argument, though it indicates that a valid argument is likely available.
I agree that there are different roles in the ecosystem. I wasn't suggesting homogeneity. I was just noting that the way we use natural resources is also important.
Stephano Mancuso might have an issue with your assumption that plants have "lower organic function." The average tree has more rootlets than a human has neurons, and they communicate with each other via at least 15 different chemical communications systems. ;-)
https://www.youtube.com/watch?v=AIfwFLDXFyQ
I think I actually assign vastly more moral weight to rocks than animals. My view on factory farming is "yeah, I'd like the conditions of animals to be improved as long as it doesn't affect my burger's price". On the other hand, I quite frequently become sad when I drive on a road cut through rock, as I ponder the permanency of the scar that'll remain long after the road is gone.
Factory farming, of course, having no effect at all upon the landscape or environment.
I can't imagine the worldview that values purely arbitrary aesthetic considerations — if not these, but instead e.g. as in an environmentalist framework, then the importance of such things lies entirely in their effect on the biosphere ("save the rainforest, xxx species die every x minutes!", not "save the rainforest, we like pretty rocks!") and hence your position would be nonsense on the face of it anyway — ...
...— purely arbitrary aesthetic considerations above living beings / suffering / conscious beings / biodiversity and biosphere / etc.
Like, I don't think I've ever actually seen even the edgiest teenager try to claim "I don't care about anything except how things look 🩸☠️". It's sort of impressive, almost, in a sad sort of way.
There is also a possibility, though this is rather a very vague idea, that consciousness is like computation power: once you have certain minimal set of abilities you have all the abilities (like simple Turing machines being able of any possible computation), though, perhaps, in a slower package.
If your definition is like mine, not quite. Consicousness is essentially a sensory function, not an implementive function, so consciousness in and of itself can't enable anything. You need to combine it with an action mechanism. Just knowing that your toe hurts doesn't do anything for you (say you've stubbed your bare foot), you've got to wash it and bandage it. The sensation, the consciousness, lets you know that you need to do that.
Similarly a Turing machine can do any possible computation, but it can't fix your breakfast of smell the smoke in the air from the burnt toast. It's complete within a restricted domain.
It is possible you are right. However, self-consciousness may be a necessary side effect of certain level of integration of brain functions, and the ability to self-model (sorry for being vague). So I do not think consciousness can be separated from "action" functions of the brain.
What would it even mean to have emotions without consciousness? You can have automatic danger avoidance reactions, its., but I thought emotions presuppose some kind of conscious experience
I agree, emotions presuppose consciousness.
Another thing that I would associate with consciousness is the "play" behavior. Like with everything else it is easy to misinterpret, and caution is warranted not to leap to conclusions. And here is some recent research which made headlines about bumblebees playing with balls (apparently young bumblebees were more playful):
https://www.science.org/content/article/are-these-bumble-bees-playing-toys
What is consciousness?
I have this question too. It seems to me the experience/concept people on this blog refer to as consciousness I would call self-consciousness.
I see no reason to doubt that bees feel emotions, and they are definitely intelligent to at least some degree. Whether they're conscious or not is going to depend on your precise definition of consciousness. They meet my definition, but I tend to shade in the direction of panpsychism, i.e. I consider an air conditioner to be minimally conscious. (If you disagree, it's probably because we differ on exactly what "conscious" means rather than disagreeing about anything substantive.)
I loved this. Is that rational?
I always thought that before we try emulating human intelligence, we should try studying and emulating insects. They are far, far simpler (and hence potentially easier to understand and emulate than higher animals) but still capable of astonishingly complex behavior. And in many practical applications insect intelligence would be more than sufficient. Think of self driving cars or autonomous delivery systems: putting an equivalent of a bee brain to steer your car would be amazing.
I am not do sure for the self driving car example: small animals seem agile, but it's mostly because of their immense strenght to weight ratio and body resistance, and also speed of mouvements that hide the details to human vision: when looked at in macro and slowed down, fast insects like bees or ants looks clumsy and bump in many obstacles, just rebounding.
Big animals like a human or, even worse, elephant, move much more accurately and carefully cause they operate in a regime where bumps or imbalance result in injuries. Cars are much more like the later, so would not really be a good fit for fast, approximate "let's try until success" control strategy.
A very good observation, I might have been too carried by my enthusiasm. But in any case, I did not mean copying exactly bee brain to pilot your car, and there are ways to overcome what you pointed out above: for example, you do not want cars to bump against one another as a part of your movement algorithm, but having 360deg LIDAR permits you to bump virtually, which is quite acceptable. Secondly, roads are normally quite predictable, and you can adapt insect-like algorithms to utilize this predictability and give equally predictable reactions. Note that bees live in a very complex, unpredictable environment, and have to move very fast in 3d: simple randomized approaches may work better in this case than more complex intelligence. When things go bad on the road (some accident), the simple but fast insect-like brain might come in handy, and perform better than more sophisticated approaches. Of course it may happen that in case of self-driving cars you really need high-resolution computer vision in which case this application of insect brains is off, but from engineering perspective, insect cognition which does so much with so little is very attractive.
Small(ish) NN similar to insect brains are certainly powerful, and would offer good preformances especially if hard-coded in specialized processors. Afaik self driving cars are already using this approach at least partially, they use NN trained on stimulated and actual roads.
My impression is that self driving is mostly solved in the sense it works in the vast majority of case, probably as well or better as the average human driver.
But it's not 100%, and I don't think anyone is willing to take responsability for the rare cases it fails. Constructors don't, passengers neither (who want to be passenger with the legal responsabilities of a driver? A passenger that should be able to take back control at any time, so can not do anything but passively watch the road? I don't ).
so it's kind of an ill posed problem, and even the case where everybody (including myself) thought the cake was cooked (road freight, trucks and such) is in fact not clear: humans do a lot more than driving (Paperwork, legal responsability, deterrent to stealing,....) and their salary is not huge. Driverless trucks are not able to compete when you take that into account.
Moving from A to B with a car is imho already solved by current tech... But it seems human drivers do a lot more, and even if they kind of suck to safely go from A to B, making them simple passengers proved hard in practice.
By the way, the critical issue of responsability for accidents makes NN or other similar techniques inherently problematic: for classic programming (procedural or variants) you can trace (in principle) the reason for some action to specific part of the algorithm, maybe even actual lines of code. The system is deterministic and can be analysed using classic reductionism (Mostly. Bugs and poorly written software already makes this very hard sometimes), and this is very convenient to establish responsabilities...
NN are still determinist (or could be, even if i think many include randomness because it often improve outputs) but can not really be analysed in the same reductionist way. So what if it fails? A bug? A exceptional circumstance that could not have been anticipated? Insufficient training? Who pay for it? Who get fired?
This is why there is some interest in neuro-symbolic/explainable AI: it still uses NN, but with a structure which is (sort of) understandable to humans: In some cases it means that NN encodes a sentence/set of sentences in fuzzy logic (it may be a very large set of sentences, and the logic is, well, fuzzy, but it is still better than conventional NN in terms of analyzing what AI does. I am not sure there are any attempts to apply neurosymbolic computation to self-driving cars, but it is certainly applicable in others (like expert systems judging people, where you can actually discover a bias). My best guess is that the future robotic systems (like self-driving vehicles) will have some instinctive component based on traditional NN (maybe inspired by social insect brains) and some symbolic component encoding moral, legal, and physical rules and getting veto.
The city of San Francisco doubts that it's as competent as the average driver. It tends to get into stalls in bus lanes and won't get out of the way of emergency vehicles.
> putting an equivalent of a bee brain to steer your car would be amazing
Now I'm just imagining someone turning on a giant bug light and all the cars suddenly start driving towards it.
So far they're still trying to emulate worms with brains far smaller than the ones bees have, and they haven't gotten it working correctly yet.
I know a colony of mrówki (Polish ants) that are in fact quite intelligent. We share information during the months in which the mrówki are active.
I have often wondered about alien life. If we ever met one, would we recognize a space alien as "intelligent" or even as alive st all?
I guess eventually yes, especially in case of space alien who created sufficiently advanced civilization, though some victims of initial misunderstanding are probably inevitable. Which, unfortunately applies also to aliens visiting the Earth, and that is one more reason not to abandon technical progress and get to them before they get to us
'I have often wondered about alien life. If we ever met one, would we recognize a space alien as "intelligent"'...
The recent advances in AI made me think about this as well. I'm no longer so convinced that humans are even the smartest creatures on this planet... if there is some living thing that's much smarter than us (a massive fungus brain?), how would we even know? I don't think dogs have a very clear understanding that we're "smarter" than them. If something out there is doing inscrutable hyperintelligent things, people probably wouldn't even notice.
It's the mice.
More seriously, how do you define "intelligent". If you're going to say that we aren't the most intelligent creatures on the planet, you need a definition that can be tested against. Clearly most of the ones we use are human-centric, but what else would you propose?
I think I have a fairly general definition that would work: intelligence is the ability to identify and make use of similarities between different things, at increasing levels of abstraction.
Like, geometry starts when you notice that a round stone and a round tree trunk have something in common. Management/psychology start when you notice that different people react in similar ways to you whacking them, and you use this to obtain compliance from more people. And the ability to recognise that the marks-on-paper I love you are the same as the waves-in-air je t’adore is several levels of abstraction higher, and indicates more intelligence.
I don't really think that would work, unless you consider all patterns detected equally valuable. Noticing that a tree branch has a circular cross section is a lot easier for a person than for a rhinoceros. We've got grasping members specifically designed to grab small cylindrical objects which we inherited from our tree-dwelling ancestors. An octopus would be more likely to notice how well a sucker adhered to it, or what the surface color variation was...which you didn't even consider significant enough to mention. (And all patterns would have to include echos and UV coloring and, oh, smells, etc. Those are what some species consider important patterns.)
"...consider all patterns detected equally valuable."
In terms of detection, I think we do have to say that, because we have no way to prejudge which patterns are going to turn out to be valuable, and which aren't. But my definition does say "identify and make use of" - so if some individuals or some group seem to identify useless patterns, and others seem to identify useful patterns and make use of them, then we could conclude that the latter group is more intelligent.
As for the other variations you mention... a truly general definition has to work across all of those difference. If octopuses are very good at observing sucker-related patterns, and on that basis develop a sucker-based mathematics, then we can safely say that they're intelligent.
You share information with the ants?
They also share with me. You know, there's a new dog a little ways from here and it might not be friendly.
Or, in my case, someone dropped a sugar cookie and you might want to take advantage of this.
Fascinating! How do you communicate with them? Specifically, how would you/they exchange those two messages?
Twitches of the tail or antennas, body posture, the angle of the whiskers, the usual things.
Speaking of communication, do you prefer the Polish word because it sounds a little similar to a Joycean mrkgnao?
No, because these mrówki are from Poland.
I realize this isn't a problem for everyone, but the style of this was very distracting. I could practically feel the unnecessary unrelated meme pics every third paragraph break. And the unexpected spider prominently on display immediately when loading the page, while relevant I guess, didn't exactly give me a warm fuzzy feeling to start reading with.
Agreed. I could handle one or two "funny"/"wacky" jokes, but one per paragraph is definitely too much.
It was fine for me. But then I'm middle/ low class. I love sports. Bring on the slapstick.
It's understandable if you found the style distracting, but I'm confused what you mean by "unnecessary unrelated meme pics". Aside from the spider at the top, aren't all the pictures directly related to what's being described in the text? And (IMO at least) none of them are memes
There exists a style of blog posts, in which the flow of text is frequently interrupted by "meme pics" which are related to the content of the preceeding paragraph or sentence, and which are supposedly intended to lighten up the mood of the reader. This style was somewhat popular a couple of years ago and has thankfully fallen out of fashion now.
"Maybe later" meant that they were reminded of that blogging style.
The prose is right on the edge, for me. I very much respect the Feynmanesque decision to keep the essay conversational and jargon-free, but a few of the puns might have been better left off.
The subject matter is fascinating, though, and I did pick up the book.
Agreed. I generally found it fun and readable, but thought a few bits of silliness were laid on *slightly* too thick.
In general, I agree. I think most of the community reviews could use a bit more of this type of humor, but this one in particular used too much. It actually got in the way of my comprehension.
Still, I respect the effort. :-)
Yeah, I had to stop for awhile and figure out the Theabee line - I couldn't figure out what part of the experimental setup that was, and took a bit to realize it was a joke.
A bit easier after that, after I adjusted to the style, but a bit distracting at first.
Yeah, it was too much, but toning it down a little would make it perfect.
Wow, better avoid ACX posts then. And most of its blogroll.
I personally found the writing style quite engaging and fun
Now that you mention it, I did feel a little jolted, though the intrinsic interest of the material carried me through.
I'm taking the opposite position, I quite enjoyed it
Put me on "Team Distracting".
I was also not very fond of the style at all.
What is it like to bee?
Or not to bee?
What is it like to bee a bat?
Not batty.
I have read several of the great bee books (like Honeybee Democracy), and the topic gets better and more remarkable as one moves along. Thank you for this review. I loved it. I suspect that most readers will stop in the middle, but for those who continue, the end material is the best of all.
Oh, is there a Bee book that is similar to E.O. Wilsons tome, "The Ants"?
You left out the fact that the scientific study of bees is called "melittology".
And "melissa" means honeybee.
"There is no such book for humans because chapter 37 will be blank."
So, if we or any entity ever learns to predict humans in this way, our entire species ceases to have ever been conscious?
Humans are already quite predictable. Otherwise complex societies would be impossible. Of course quite predictable does not imply completely predictable, but neither are insects.
Yeah. I found that statement very odd. Humans are quite predictable and we predict other humans and their reactions, in aggregate or at an individual level, quite accurately, just not perfectly.
There’s no way we could safely operate tons of metal at 60 mph if we weren’t, in at least some conditions, extremely predictable. Air travel emphasizes this in a different way - you wouldn’t get in a metal tube with 300 strangers at 30,000 feet for six hours if you didn’t predict they would behave basically the way they actually do.
Mad Max: Above Thunderdome.
It would just help people realize that consciousness is an incoherent concept, and that we're already able to answer for any given things whether it contains any of the real subcomponents of consciousness, like ability to learn.
So we are able to answer whether anything has qualia, or moral weight?
About qualia? Not now, and probably never.
About moral weight? First, science alone cannot answer value questions. Moral weight depends on your ethical assumptions. But of course science can (potentially) answer if those assumptions are satisfied in particular cases. In this case, sometimes it seems we can and we do. The research underlying the book "Consciousness And The Brain" in a last year's entry into book review contest:
https://astralcodexten.substack.com/p/your-book-review-consciousness-and
is apparently useful as a factor in deciding the fate of people in deep coma. It can also be used to debunk the claims of Peter Singer about morality of killing born babies (they are probably conscious, contrary to the claims by P.Singer).
So, as pointed out in the aforementioned review (as well as many other posts on this blog) we already know or suspect quite a bit about the role of consciousness and its underlying neural mechanisms. Of course we are no closer than ever to solving the "hard" problem of consciousness because it is probably unsolvable. I find the following analogy as useful when thinking about hard problem of consciousness: Once the question about the true nature of matter was an important part of philosophy. Then came physics, and now we know the structure of matter to the teeny tiny bits, and about the forces which hold it together. But did it solve the real "hard" problem of the nature of matter? I am no philosopher, but I guess the answer is no. However, aside from some fanatic philosopher here and there nobody cares anymore: our knowledge about how the matter operates, and how to manipulate it, and curiosity about what we do not know yet in this regard displaced almost completely the interest in the "hard" problem of matter. I believe the same will happen with the "hard" problem of consciousness: it will die natural death when our knowledge about practicalities of consciousness become deep enough.
“I believe the same will happen with the "hard" problem of consciousness: it will die natural death when our knowledge about practicalities of consciousness become deep enough.”
This is exactly the argument of Anil Seth in his book Being You.
He describes what he calls the “real problem” of consciousness, which is “to explain, predict, and control the phenomenological properties of conscious experience.”
In "What It Is Like To Be A Bat" by Thomas Nagel, proposes that we can never know what it is like to be a bat, because we didn't develop our mental capacities as bats. To know what it is like, you have to actually be a bat. Interestingly, I believe that same thing, more or less, is true for individual people.
Yeah, I didn't like this line also. Perhaps humans are more a difference in amount, and not in 'kind' of intelligence
Note that it's a quote from David Deutsch, and the review is largely arguing against it.
Yep. Or rather, whatever it is will come up with a new name for what we humans do, and will reserve "consciousness" for whatever ineffable processes are exclusive to itself.
Yeah that seems like a wild thing especially to say to an economist when predicting human behavior is so much of economics.
Hard to ask if things are conscious when no one can define consciousness in a coherent purely materialist way. Just like free will, people hallucinate that there must be consciousness as a result of the instinct to reify the nonexistent center of a cluster of correlated factors described by words. But there's nothing at the center of the meaning cloud the word "consciousness" points to, just a blank space at the weighted average center point of n-dimensional space where the subconcepts we associate with consciousness cluster around. The question is meaningless because the concept being asked about is incoherent.
Still, I'd read a book about bee neurology, seems like a cool topic of study. The review was written in a funny and punchy way, although there were times I was confused about exactly what the book was claiming or what had happened in one experiment or another. I wanted a deeper exploration of e.g. wtf is going on with bee dances, and whether/why they don't matter?
Agreed w/ your second paragraph -- there were a few instances where I didn't understand what was being described as well as I would've liked.
Aside from the bee dances that you mention, I didn't quite get what's up with the honeycomb angles:
"But they don’t build them purely vertically - they keep it at an angle so that the honey’s viscosity and adhesion keep it inside the vessel. The cells are tilted slightly downwards."
If they were worried about honey staying inside the vessel, why not tilt them slightly upwards?
I think that's what he means. The "inside" is tilted downwards = the outside edge is tilted upwards.
The bigger issue is that our prevailing materialist ontology isn't rich enough to describe consciousness even in principle. With a richer ontology, we could argue more sensibly about whether it's real or not. "Materialists" shouting that consciousness can't be real because it doesn't correspond to their concepts do themselves no favours.
I'm almost certain that whatever thing or cluster of things we call consciousness will turn out to have a material basis, but the current debate is so bad that the people who believe in souls look like the scientific ones.
>But there's nothing at the center of the meaning cloud the word "consciousness" points to, just a blank space at the weighted average center point of n-dimensional space where the subconcepts we associate with consciousness cluster around
Yes, "Consciousness" has too many meanings, not too few. Concerns about the mirror test, as mentioned in the book an review, point to the sense-of-self definition., Concerns about emotion point to phenomenality or qualia. It's likely you can have the one without the other.
But in what way is it a gotcha to point out that "consciousness" the word is polysemous? If any of the subconcepts is problematic, then you still have a problem.
Phenomenality and qualia are also meaningless. No one can define them in a way they are measurable or falsifiable.
Second sentence is true, first sentence doesn't follow.
How can something have meaning as a concept if it can't be defined in a way that you can interact with?
I can interact with my own qualia just fine.
And I exercise my free will every day.
I've got to disagree. I have a good definition of consciousness that is coherent and materialistic. But it has the result that I consider an air conditioner to be conscious at a minimal level, and most people won't accept that. I assume that other people also have coherent definitions of consciousness, though many can't verbalize them. The problem is that we don't AGREE on a definition of consciousness.
Since my definition is clearly an idiosyncrasy, I would be willing to accept many alternative definitions, but people generally just hand wave.
This seems the right context to make my definition explicit:
Consciousness is the ability to perceive some portion of reality and react appropriately.
I'm a bit dubious about the word "reality" in that definition, and the reaction isn't actually a part of consciousness, but is needed in order for the consciousness to be detected.
I assume 'appropriately' stands for something like 'in an ordered way, which another consciousness can, with effort, perceive as evidence of intent'?
You did acknowledge the caveats, but it's difficult to see how this definition excludes a lot of simple, non-volitional chemistry - or how it reliably includes things like dream-states.
Well, I don't equate consciousness with volition. And it intentionally includes lots of non-volitional actions. When I neuron responds to a stimulus by opening a sodium channel, I consider that conscious action at a minimal level. Volition would imply that it had a model of what it was doing and selected that response, but consciousness is not dependent on there being a response, it's just that without a response you can't detect it.
"Appropriately" means pretty much "in an ordered way", but note that the response is only how we detect the consciousness, not the consciousness in-and-of itself. Consciousness is a sensory thing, not an effector thing. Sometimes what it senses are internal states, and we call that self-consciousness, or sometimes awareness.
Again, this is my idiosyncratic definition. If someone makes another operational definition, I'll be willing to use that when talking to them. I made this one when trying to figure out what it meant for an AI to be conscious. (And, of course, whether it acts on its consciousness depends on how closely you look. It will definitely at least flip bit states around...which is a sort of action...so this isn't a totally bounded definition.)
I think your definition is too broad to represent what is usually meant by conciousness.
My definition would be more like: Internal awareness on the ability to percieve and react to reality.
Self-awareness (as tested in a mirror test) would not necessary for conciousness, but conciousness is necessary for self-awareness.
It may be useful to distinguish between consciousness of one's surroundings, and the objects occupying them, versus consciousness of one's mental self.
Obviously all animals possess the first, but I'm not so sure the second is well developed in most animals, even larger animals such as reptiles and many mammals. It may be present up to a point in social animals because there are big advantages in being able to "put one's self in the other guy's shoes", to anticipate motives and thus likely actions, and that ability can then be used in reflective mode to be aware of one's own mental workings.
I think reptiles' consciousness probably, and amphibians' certainly, is most likely what we would experience as a dream like state, in which things just "happen" and drift along in a haphazard and not always consistent sequence, and there is usually less self-reflection than a fully awake state.
That is because I think dreams originally evolved in the first amphibians as a crude "record and playback" memory mechanism to allow them to retrace their path on land back to water. (Notice that dreams are most vivid when one is uncomfortable, for example too hot or cold, and especially when thirsty!) Then they were repurposed over evolutionary time to serve other functions.
> It may be useful to distinguish between consciousness of one's surroundings, and the objects occupying them, versus consciousness of one's mental self.
I think this is a useful distinction to make . A lot of things that are being ascribed to consciousness, I would consider sentience. Conscious. Sentient. Are they merely synonyms? They can be I suppose, but it seems like a missed opportunity.
<i>Just like free will, people hallucinate that there must be consciousness as a result of the instinct to reify the nonexistent center of a cluster of correlated factors described by words.</i>
Except you can't have a hallucination without being conscious...
There seems to be an endnote that doesn't link to anything. François2
That should be read as "Francois squared" - there were two Francoises and one Marie-Aimee involved.
Yeah, that took me a few seconds. It's a joke about there being two of them.
Great! I bought the book. Thank you. I loved all of it. (and got to learn several new words)
Oh and Humblebees, this is so pooh bear (A.A. Milne) that I'm calling them that from now on... bring back humblebee!
That was a great anecdote.
How well do the "bee intelligence" studies replicate? Any researcher of "bee intelligence" would be _highly_ motivated to find patterns that don't actually exist and ascribe higher level of intelligence to bees, because their own career depends on it.
It's kind of like that story about chimpanzees learning sign language, where there was a famous study showing that they can supposedly learn dozens of signs, but it failed to replicate. So by default I'd assume that the book is sorta bogus.
Using the alien example is helpful. Aliens grabbed a random human, who happened to walking on the campus of CalTech, and were able to teach him a few advanced alien mathematical concepts. But when other alien researches grabbed another random human they weren’t able to teach him at all.
TLDR maybe it was a really smart chimp.
I thought I knew a thing or two about bees, but there was a lot of new stuff for me in this review. So thanks for that!
Regarding bees in space -- "The bees got rid of the slight angle downwards - there’s no gravity in space, and thus no need for the angle." -- Surely a bigger reason is that there is no 'down' in space so they couldn't build downwards even if they wanted to just for fun.
And last but not least: the parallels between the bees' honeycomb building behavior as described in the review and the way that bioelectric fields control morphogenesis in multicellular organisms (see Michael Levin et al, and/or videos such as https://www.youtube.com/watch?v=K5VI0u5_12k ) are striking! I suspect that there's some very interesting work that might be done in this area!
In space, don't humans still naturally orient to a shared up/down orientation? I'd suspect that a sign of 'higher' cognition would actually be getting fooled into believing there's an up/down orientation outside of the direct signal.
That's an interesting suggestion. But in the absence of artificial gravity, do humans in space really orient to an agreed up/down orientation? TBH it's not what I'd have expected. I mean, 'down' would still be where the earth is, but since that would change from one moment to the next with the ship's drifting orientation, I don't think that would be a very practical 'anchor'. I'd have expected something more local, case specific, and concrete, along the lines of the Honolulu residents' system of mauka/makai and Ewa/Diamond Head -- but with 5 or 6 dimensions (eg nose/stern, navigation bay/food dispenser/power panel) instead of the usual cardinal 4.
I was thinking in simpler terms. In the absence of directions defined by gravity, they would be defined by familiar features (or an arbitrarily imposed definition), where one surface would be designed the "floor" and another the "ceiling".
I think you see this tendency even in the presence of gravity. I went to an illusion museum in NYC once, where they have rooms made up to look upside-down or skewed. Despite gravity giving you signals about the true orientation of the world, it's difficult to keep your brain from defining a different orientation.
If bees are NOT susceptible to that kind of thing, it suggests a lower-order organization along strict signals like gravity, not that there are many layers in between.
TBH I'm not sure how spaceships like the ISS are actually arranged in terms of whether or not there would be one particular internal surface that would look more like eg a floor than the rest of the interior -- maybe the stern would be more floor-like? I dunno. I'm also not sure whether it would be helpful or actually even more disorienting to use the usual conventional earth-bound this-way-down cues that we use unthinkingly inside just about every room we've ever been in -- outside of a fair-ground 'wildhouse' or a NYC museum. In my case, I think the conflicting cues would probably make me instantly nauseous, although, like sailors, I imagine astronauts are probably trained and well-practiced in coping with that kind of problematic sensory input.
When it comes to bees in zero-g, I'd expect them to also be dealing with conflicting sensory inputs, but of a different kind, eg light polarization as opposed to the orientation of visual objects -- which in turn would open up a whole new class of visual/sensory illusions. So I think it might be hard to tease apart the effect (if any) of lower vs higher levels of organization. It would be an interesting challenge to devise such an experiment though!
I loved the joky style. It gave my younger son another reason to be interested in biology as a career: the chance that you might spend your days in the lab making robot spiders fight bumblebees.
Or RoboBees... https://wyss.harvard.edu/technology/robobees-autonomous-flying-microrobots/
Erik Hoel, is that you? Excellent review; best so far.
The hexagonal shape is highly useful, but bees may "just" build them round and then physics take over: the wax gets cooler and snaps into form. As in lava-stones: https://www.geologyin.com/2015/10/mystery-solved-how-these-rocks-got.html
That was my suspicion. If you take a lot of flexible roundish things of the same size and cram them together under enough pressure, you tend to wind up with hexagons.
So whence the round shape?
The same reason bubbles are spherical? It's the most efficient shape for surface-to-volume ratio.
Why is there a physics explanation for hexagonal and round?
I'm not clear on what you're asking. Could you rephrase, perhaps at more length?
Physics appears to predict both round and hexagonal holes from the same starting conditions.
I don't think this is the case?
it depends on the surroundings I believe. An individual bubble will be round, but if you surround it by other bubbles, you will notice their bases tend to take on a more hexagonal shape. When there's equal air pressure, you're going to get a straight line (hence a polygon), and layout that looks like hexagonal tiling packs spheres in the most efficient way on a 2D surface.
The review:
A quick (relatively) and fun (absolutely) read. I liked the overall tone. I have two main criticisms.
First, it didn't make its thesis clear. I think it was using a book about bee intelligence as a springboard for considering AI consciousness. But it lacked enough discussion of what "intelligence" and "consciousness" are, and especially the sorts of entities that they can apply to. (Are individual bees intelligent, or is their intelligent behavior the result of interactions as a hive, and is therefore the hive the source of "intelligence"? And what does that say about humans and tribes and cities and civilizations and the Internet? And about AIs?) The mention of "consciousness" without an attempt at a definition particular bugs me, although that may just be a pet peeve.
Second, the humor interfered with the content. I liked the humor, as humor, and while I think the amount was on the high side, on reflection it seems within tolerances, and it wasn't the amount itself that caused problems. Rather, the problem was the times where there was a paragraph that seemed to be leading to a point, and then the last sentence or two was a joke, and then the next paragraph moved on and picked up a new thread as if the point had been made. Leaving me wondering where it was going, and creating a disjointed feeling throughout the review. And maybe I'm sensitive to this because I suspect I do too much of it myself, and so I'm, well, not strictly hypocritical, but close enough. If I weren't trying to write mini-reviews of all the reviews to aid in my later voting (and to try to get myself to pay more attention and think more rigorously about the subjects), I might not even mention it. But then, it did interfere with my comprehension, so.
The book:
I'm not entirely sure what the book was about, other than lots of anecdotes about bee brains and the cool things they can do. There are some pop-sci type books that are basically a survey of a field at a point in time, and maybe this is one. Maybe the author wasn't trying to force a message onto data that is as yet too incomplete to sustain the load. That'd be nice, for a change.
Overall, it put me in mind of a long article that I recently read, about the Argentine ant (Linepithema humile) and how it was forming trans-continental ethno-nationalist super-colonies.
"You know squirrels bury nuts so they can dig them up later. Well, some people did a very cruel experiment. They put a squirrel, given some nuts or something (I don’t know how they set up the experiment), on a concrete floor. The squirrel did exactly the same behavior with its hind legs with the nuts and put the nuts there and so on. Even though it was having no effect whatsoever. We see the point of scrabbling with your hind legs and then nudging the nuts over there and so on, but it doesn’t. It’s just a program being enacted by its genes."
---
"You know humans reproduce sexually so they can pass on their genes. Well some alien scientists did a very cruel experiment. They put a human (who wanted to have sex) in front of some fake images of an attractive partner. The human did exactly the same behaviour -- looking at the images, getting aroused and so on. Even though it was having no effect whatsoever. We see the point of sexual arousal and so on, but it doesn’t. It’s just a program being enacted by its genes."
---
My point is, I don't think we have a leg to stand on. Lots of our behaviour comes from evolution just programming vague 'drives' into us, which we then turn to our own ends. Even if squirrels just act out 'burying nuts' on a concrete floor, that doesn't mean they are mindless. Maybe their little squirrel minds just get a kick out of scrabbling at the floor and placing a nut there, and they are acting perfectly rationally given the drive that evolution has installed. Yes, it's not actually achieving the goal, but nor are we when we look at porn, eat junk food, etc. Are we mindless?
I'm sure squirrels, like a lot of animals, including primitive humans, are into sympathetic magic, i.e. enacting symbolic imitations of events they would like to happen, such as a rain dance with water sprinkled around in a quest for rain.
I was sitting on a park bench once, eating some nuts, and a squirrel approached and stopped nearby on the ground and started miming eating a nut while staring intently at me. That seemed like an example of sympathetic magic, and quite plausibly a conscious decision on its part to play act at eating to try and convince me to share some nuts with it! But then it could just have been an innate response, physically playing out a longing or instinct going through its head no different to the squirrel scratching concrete
There are even child (well, chick) sacrifices by birds to gods (alligators in this case):
https://eu.jacksonville.com/story/news/2016/03/02/everglades-birds-sacrifice-young-alligators-protection-study-says/15702230007/
Similarly, I took down my bird feeder for cleaning and didn't get around to putting it back up for days. I was standing at the window where the feeder had been in view and 5 or so house sparrows were in a bush and one flew up to where the bird feeder had been and hoovered for a second - and I could have sworn looked over at me critically! I dismissed it as anthropomorphizing and then it happened again six months later when I ran out of seed. FWIW
I have seen dogs do what I have interpreted to be the same kind of thing, try to bury a toy or a bone under a surface that can’t be buried into.
Or maybe it's communication?
One of the more reliable ways to convince cats to sit on my lap is to hold my hands over my lap and make kneading/petting motions. Is this communication? Sympathetic magic? Stimulation of reflexes? I have no idea. It just seems to work more often than other stuff, so I keep doing it. :-)
I also find the framing here to be questionable. I would have thought it was commonsense that all behavior lies on a continuum from mindless program to highly complex (which has nothing whatever to do with "consciousness" whatever that is). I doubt any behavior in any animal is entirely mindless (that is, it uses exactly zero information from the outside environment), and that includes the bacteria. I doubt equally that there are any behaviors that are entirely free of genetic influences (including this post I am typing). Really, we need better operational definitions than a binary variable.
Found the numerous stretched attempts at wordplay humour to be...well...a buzzkill, distacting from the actual content of the review. Which wasn't the easiest juice squeeze, either; I kept wavering between thinking the central theme was cognition in bees specifically, or the meta-question of how to define consciousness writ large, as illustrated by bees. (And how much of each was book vs. review-of-book.) Arbitrary-seeming anecdotes and tangents further undermine focus, such as the overdetermined obligatory what the refrance to AI. Also feels like there's a missing paragraph(s) or something, the ending is just kinda abrupt.
I dunno, it mostly just felt like reading a draft in need of further editing. Potentially good material that could be tightened up into a solid review, probably.
"Note that there are lessons for specialisation and other forms of social organisation here - if you are more sensitive to noxious smells, you spend more time fanning, and you get better at it. This may cement differences in job allocation. Small differences in preference become permanent differences between individuals. This sort of thing probably happens in humans - ..."
This strikes me as super interesting. I am currently reading the WEIRD-book, and there is a very similar passage in the development of the Big 5 personality traits, which are seemingly only found in actually WEIRD or WEIRD-like populations. In more "primitive" tribes, where people are generalists, the Big 5 do not emerge at all. Instead fewer, down to even just two traits are detectable, and they don't necessarily line up with any Big5.
Expressed differently: Big5 intercorrelations are a spectrum, and the WEIRD populations have the clearest distinction between them. The resulting individual specialisation is supposed to be one of the major factors of success of The West, enabled in the first place by a social order that does not rely anymore on kinship relations. Instead, people are encouraged to specialize and partake in voluntary associations of their own choosing, competing in smaller groups or even just for themselves.
I want to chime in that I liked the jokes. I am often too serious myself and therefore I appreciate it very much when others author introduces some victimless shenanigans. For me, those jokes are independent of the content quality. Who doesn't like puns anyway?
> In more "primitive" tribes, where people are generalists, the Big 5 do not emerge at all. Instead fewer, down to even just two traits are detectable, and they don't necessarily line up with any Big5.
I'd put my money on this result failing to replicate...
To be fair, Joseph Henrich specifically and clearly wrote that this particular chapter is by far the most speculative of the book, because the supporting studies are relatively new and rare. But it is still a number of studies, not a single one. And they all have a theme going-on: Cross-cultural Big5 studies all find higher correlation between Big5 traits than in Western studies. But he also points out in his favor, that almost all cross-cultural studies are done with undergrad students in the respective countries, i.e. a western- or "WEIRD"-biased selection of people.
So you might still get your money back, but I wouldn't be as confident as you seem to be.
I'm not as confident about this as I (possibly) seem to be. I freely admit I know almost nothing about this specific subject. Just wanted to point out that this result goes against (my) intuition, and social studies with counter-intuitive results have a history of failing to replicate. Your response pushed me slightly towards "might replicate" because you say there are a number of studies - but not much, as the author himself said it is speculative, and the results are recent.
> whether they become pollen or nectar foragers
I thought the bees themselves only wanted the nectar, and that pollen was an incidental the flowers stuck on them after attracting them with nectar.
Oh no, bees absolutely eat pollen. It's how they get protein and fat - nectar doesn't really contain either.
Isn't Gary Taube's whole thing that you get fat by eating carbs rather than fat? Fair point about protein though.
If you somehow ate a diet with absolutely zero fat, you would start having health problems due to essential fatty acid deficiency. Bees are obviously quite different from us, but they can't do without a dietary fat source either.
Nice review! I actually gave a speech on this recently. In Chittka’s lab, a study was done to try and see if bees would play with small wooden balls when there was seemingly no real incentive to do so, and it was found that they did, which I think is another point for consciousness/emotion. Of course, there’s always the conversation of “was this really play” and whatnot, but the researchers I’d say justify it pretty well. I’m glad to see the topic getting some attention here!
https://www.scientificamerican.com/article/ball-rolling-bumble-bees-just-wanna-have-fun/#
This was an interesting, fun and informative read. Kudos to the writer!
Re: "Bees exist in that great hinterland of consciousness"
Also plants, somewhat:
Bees exist in that great hinterland of consciousness, Your brain is not the root of cognition.
--Nautilus, March 7, 2023
Also there's a book "The Hidden Life of Trees" which Tim Urban is reading per Twitter.
Based on the one following quote I recommend it(and could someone review as I am backlogged). : )
“If a tree falls in the forest there are other trees listening.”
The squirrel study is the German paper 'Eibl-Eibesfeldt, Irenaues 1963. "Angeborenes und Erworbenes im Verhalten einiger Säuger". _Zeitschrift für Tierpsychologie_, volume 20, pg705-754' https://gwern.net/doc/psychology/animal/1963-eibleibesfeldt.pdf
I found it by the obvious Google Scholar search of "squirrel burying instinct experiment concrete floor", where the 6th hit (https://link.springer.com/chapter/10.1007/978-1-4899-4656-0_12) has the highlighted excerpt "the squirrels were capable of 'burying' a nut on the bare floor in the"; while not specifying 'concrete', this is clearly very relevant, and looking at it in Libgen, it is clearly the relevant research, and the date is correct for Deutsch to be very distantly recalling it. It cites 'Eibl-Eibesfeldt 1963', which gave me conniptions as nothing whatsoever was mentioning it and the one other citation simply failed to actually give a reference for what 'Eibl-Eibesfeldt 1963' *was* and even the E-E GS profile wasn't turning up anything relevant for '1963', until I went back to the original hit and specifically downloaded its endnote/reference section and realized that there was nothing in GS because it was a *German-language* paper. I can't read German, but the paper does have photos of squirrels & nuts, so I assume it's the right one.
You probably don't hear this enough, so: thank you for doing this sort of thing!
An English summary of the experiment: Squirrel infants were taken from their nest while still blind. They were raised in individual cages and given liquid food. At ~2 months old, the researchers opened the cage doors and gave hazelnuts to the squirrels. The squirrels ate some of the nuts and tried to bury others in the ground.
Researchers concluded that this must be a fixed and inborn skill, because the squirrels had never seen hazelnuts before, and never observed other animals burying them.
---
Should I really be translating this? It feels like a high amount of animal cruelty for unclear benefits :-/
My predictions:
1) A large fraction of these bee experiments won't replicate. Some of the bee stories always struck me as unlikely wishful thinking by researchers doing poor quality work.
2) The tool use video doesn't show the bee pulling the rope; to me, it looks like the bee is attempting to climb the flower under the glass, and doesn't realize the effect is pulling the flower. You can see the bee desperately trying to reach under the glass instead of stepping back and pulling the rope as a human would.
3) Brain waves correlate to the brain being active, not to consciousness. Nobody doubts bees have a brain.
It's very interesting to understand what bees can and cannot do, what they can and cannot learn. But whenever I read about it, I get the strong sense that the source I'm reading is biased in the bees' favor. Does the book report examples of trivial-sounding things bees cannot do? Those must exist, right? Why do sources never report them?
(For example, I once read an assessment of Terence Tao's mathematical abilities at age 8. The person assessing was very impressed and gave examples of what Tao could do at that age. But importantly, he also gave a couple of examples of easy-ish things Tao could NOT do at that age -- and that made it seem a lot more trustworthy. I want the analogous thing for bees.)
> I get the strong sense that the source I'm reading is biased in the bees' favor
Based on entomologists of my acquaintance, this seems right. Although it's quite possible that the rest of us are instead biased in mammals' favor.
Well, Humans favor. I've seen the same sort of thing (wishful research results) but involving primates and dolphins.
I'm mostly speaking for myself: I did not realize I was this much of a mammal chauvinist until speaking with an entymologist.
Your last question suggests to me that you are Irish. Or, rather, I ID the construction, "Riddle me this." as a very Irish rhetorical device indeed:
"Now, you’re supposed to be
An educated man,’
I hear him say. ‘Puzzle me
The right answer to that one.’"
- Seamus Heaney, "A Casualty"
_________________
So. Am I far from the mark or no?
I thought it was a Batman reference?
https://www.youtube.com/watch?v=HU3DMMzjVig&t=23s
Hands down my favorite review thus far. A stronger contender may yet emerge from those as yet unposted but otherwise I've found my vote.
I greatly enjoyed this review, and absolutely loved the puns and humor. I may very well end up voting for this one!
CW for crab spider was very justified.
I really loved all the puns
Re: waggle dance, bee differences and the false certainty of Big Data
"the 2008 financial crisis arose after people placed unquestioning faith in mathematically neat models of an artificially simple reality. Big data carries with it the promise of certainty, but in truth it usually provides a huge amount of information about a narrow field of knowledge.
...
There is a parallel in the behaviour of bees, which do not make the most of the system they have evolved to collect nectar and pollen. Although they have an efficient way of communicating about the direction of reliable food sources, the waggle dance, a significant proportion of the hive seems to ignore it altogether and journeys off at random. In the short term, the hive would be better off if all bees slavishly followed the waggle dance, and for a time this random behaviour baffled scientists, who wondered why 20 million years of bee evolution had not enforced a greater level of behavioural compliance. However, what they discovered was fascinating: without these rogue bees, the hive would get stuck in what complexity theorists call ‘a local maximum’; they would be so efficient at collecting food from known sources that, once these existing sources of food dried up, they wouldn’t know where to go next and the hive would starve to death. So the rogue bees are, in a sense, the hive’s research and development function, and their inefficiency pays off handsomely when they discover a fresh source of food. It is precisely because they do not concentrate exclusively on short-term efficiency that bees have survived so many million years.”
--Rory Sutherland, Alchemy, 2021
""the 2008 financial crisis arose after people placed unquestioning faith in mathematically neat models of an artificially simple reality. Big data carries with it the promise of certainty, but in truth it usually provides a huge amount of information about a narrow field of knowledge."
Actually, the '08 crisis arose because greed overcame good judgement. There were some outright lies involved too. So maybe not parallel to anything bees do.
"It is the nature of every grain of corn to become wheat and every precious metal to become gold and all procreation to lead to the procreation of the human race. Therefore a master says that there is no animal which does not possess some likeness to human beings.”
--Meister Eckhart (1260-1328)
Oh, we can easily find other parallels. The industriousness of bees, their social cohesion, they way they protect the gene lines. But do bees deceive each other? Some animals do--I've seen some evidence for higher primates.
Book review: "Almost every chapter discussed in this book in some way reflects on this question of complex cognition versus instinct."
Given humans have a primitive and higher mind can this model(T. Urban, btw) be applied to lower animals and surprisingly capable insects? Observed helpfulness being more primitive herd instinct than empathy to action.
Re: "do bees deceive each other"
I would think that's a more evolved/advanced, ah, feature. To deceive or not to... : )
"The germ of even the most complex ideas – reasoning, imagination, curiosity, aesthetic sense – was, Darwin argued, to be found within animals. Morality too. ‘Of all the differences between man and the lower animals,’ Darwin wrote, ‘the moral sense or conscience is by far the most important.’ Even so, non-human animals have the capacity for moral thinking, though that capacity is not as well developed as in humans.”
--Kenan Malik, The Quest for a Moral Compass, 2014
Lastly, there's a 90s movie with Ed Harris barely relevant to this exchange: Screamers. There's a couple lines near the end I want to quote but...spoiler.
Oh, and I forbid anyone to watch the sequel!
Re: "But do bees deceive each other? Some animals do..."
Animals evidently evolve larger brains largely due to social pressures and at some point become capable of deception. Maybe in gradient steps?
"Living in large groups means that you must have the ability to understand other people’s minds—their thoughts, beliefs, emotions, and wants—if you want to get along or get ahead. Such brainpower requires actual brain cells, which is why cerebral cortex size—the part of your brain involving thinking about the minds of others—is directly related to the size of the social groups they inhabit.”
--Nicholas Epley, Mindwise, 2014
"Working out the social politics of who you can trust and why is, quite literally, what a very large part of our brain has evolved to do.”
--Douglas Adams
I really like this idea, but I'll play devil's advocate:
Maybe we can get away with it if we have enough of a surplus, and institutions that backstop failure. Maybe the resources that we're dealing with have a huge first-mover advantage. Maybe this type of failure is exclusive to certain areas like politics, where there are extremely strong incentives for everyone to jump on the latest bandwagon in lockstep. So, what makes certain areas of finance more like politics, where it's about projecting an image and avoiding being the only one blamed, as opposed to other areas like Good Old Neighborhood Banking and Warren Buffett?
When a SF startup coder bee comes in and waggles his butt to signal "AI", there's a huge rush of other SF startup coder bees to go toward AI, not for the interest of the Hive, but for their own interests. But there's plenty of MS coder bees and Google coder bees and Amazon coder bees and open source coder bees that keep on working on whatever they're working on, maybe with a marketing drone splashing a bit of "AI" pheromone on it to keep everyone from freaking out that they're not getting in on the next big thing. This is pretty good, right? There's a sense that we may have identified an untapped resource, so the frothy surf of opportunists and under-utilized people go for it, but the depths of the ocean keep moving in their own cycles. So I'm left feeling pretty good about the tech industry.
Insightful.
It also fits with the reviewer's mentioning of "brainy bees" figuring out the rope-pully-sucrose contraption and other bees observing and doing the same. To extrapolate(speculate), at some point there would be a queue and some bees would revert to their previous pathing.
It's fascinating. I'm going to need to read the book and probably others as well!
I didn't see the relevance of Molyneux's problem, which is about blind humans who just became sighted. I imagine *sighted* humans can feel something and then recognize it. Locke's point (and Molyneux's) was that newly sighted humans haven't yet learned to associate the two senses adequately.
To be relevant, one would have to run the test with bees that had until recently been blind their whole lives, which I imagine would be hard for reasons similar to why it's hard in humans.
I think the intent was to use generalizability of skills as an indirect measure of consciousness, though I think that conflates it with "general intelligence", which I would argue is a different thing.
I question whether bees are ineffective waggle dance communicators outside of Asia(where they evolved). If they can adapt to zero gravity and get work done why not a different terrestrial environment? I realize this comparison may be apples and oranges but given what we know Bees can do...
I got the sense that in the more homogeneous abundance of Europe the selective directing of the waggle dance was simply superfluous, whereas in the ancestral environment a bee going other than to the destination indicated by the dance would strike out.
I understood that as well - from this source. Bees as a whole seem highly adaptable and seems like they could overcome the Asia to Europe move. If this is truly the case for all bees in Europe I am curious as to why. As the reviewer mentioned, funding for research can be tricky so there must be more to study.
It's not that the bees are ineffective waggle dancers in Europe, it's that the waggle dance itself is ineffective in Europe because the information it conveys is unnecessary. There's nothing for the bees to adapt to other than to stop dancing altogether, but there isn't sufficient pressure against dancing to make them stop.
I get how "diffuse light" causes a problem as a point of reference but not sure how the more "open landscape" of Europe versus the tropical forests of Asia factors in to the distance communication issue the reviewer refers to.
Since it continues to bother me I suppose I'm destined to read the book. : )
The open landscape of Europe is much more homogeneous than tropical forests in Asia. Independent of how far you go in what direction in Europe, you'll find roughly the same density of resources; following even perfect waggle dance directions doesn't improve your outcome over just picking a destination at random.
The bees use the SUN to navigate so I would think the open landscape of Europe having fewer obstacles between them and their beacon would be an easier environment to describe.
Obviously I'm missing something here.
This is the passage I'm primarily referencing:
"If you tilt the hive, the bees can adapt and continue to use the sun as a reference. But if you only give bees diffuse light, they lose their reference point. They keep waggling, but the dances become basically randomised, even with just one dancer. Other bees try to follow along, but it’s essentially a waste of time -
**they’re just getting a distance, with no direction."**
Funny review :) I come away with the impression that we've done many interesting tests on bees, and it's not the lack of tests that's the problem, but the lack of understanding of what 'consciousness' even is, and how we should test for it. Maybe it isn't even a coherent concept?
Look forward to reading more from this author when the names are revealed!
Much like "Intelligence" that way, except worse. At least with IQ we have a relatively stable metric that is probably measuring *something*, even if we don't know exactly what. With consciousness, I don't think we even have that much.
The discussion of how it's hard to show an animal can't do something reminds me a lot of LLMs
That bit about squirrels reminds me of the sphex wasps:
"The standard story has it that the female Sphex wasp will paralyze a cricket, take it to her burrow, go in to check that all is well and then come back out to drag the cricket in. So far that might sound pretty intelligent. However, if an experimenter moves the cricket a few inches while the wasp is inside, then when she emerges she will move the cricket back into place in front of the burrow and go in to check again rather than just take the cricket in directly. And she will (again, so the standard story goes) repeat this ritual over and over if the experimenter keeps moving the cricket."
https://edwardfeser.blogspot.com/2013/12/da-ya-think-im-sphexy.html
FWIW, I liked the humorous style and this is my favorite so far, although I agree that the jokes were on the high side and there were some I didn't get (e.g. Therabee and Francoise2). I also had no problem with the spider image. I was surprised to learn that so many people here can't stand looking at spider photos, since that would have never occurred to me as a problem.
Bought the book before I finished reading the intro, excellent book review
Hm. The types of qualitative analysis I am familiar with are methods of assessing text or behavior. What I am talking about are my own introspective experiences when I am in the presence of my wife. I suppose I could make videos of my behavior toward my wife and subject that to analysis, but I am not sure that would allow me or anyone else to meaningfully describe my emotions any better than I do just by feeling them.
Similarly for the color orange. You are familiar with the mental experiment involving someone who has never seen color collecting all the objective evidence they can about what color is?
Amazing writeup if pun-ishing.
The complexity in actual behavior exhibited by bees - in reaction to people being dicks - shows just how much capability that evolution has built even for the humble bee. To think that this level of complexity can be replicated from conception to design to training set to execution, then evaluated over millions of generations reproducing in a wide variety of environments with the most brutal and objective testing imaginable - reality - just shows how naive the AGI folks predicting success in a mere decade (or ten) are.
The writing style did throw me off. The topic is extremely interesting to me, I expected reading about what the book conveys on it. The prose really threw me off. I had to re-read at a couple of spots to leave out the joke and figure out what you are actually saying. IMO, it comes off as trying too hard to write like Scott does. I imagine a ChatGPT writing prompt of " Write a review for this book in the style of Scott Alexander incorporating jokes in the prose for which he is famous for "
FWIW, I got sufficiently frustrated to actually go read the book instead of finishing the review.
(The spider is kinda cute, loved it)
Happy to have learned more things about bees, and when the authors are revealed I'd like to know if there's a blog where I can read more article by this person.
For people wanting even more fun facts about bees: female bees have twice as many chromosomes as male bees (haplodiploidy).
This one just buzzes from pun to pun. I have half a mind to record an audio version for fun.
A very interesting and informative review, but also incredibly annoying. So many things could have been deleted to make the review better. To sum up, the reviewer should have avoided facetiousness, puns, supposedly clever remarks, and irrelevant references. Here are some things that should not have been there:
"I remember when I was randomly attacked by robotic crab spiders for a day, and I didn’t enjoy it much. The bees shared my opinion.
So I suppose we owe a small debt to that nasty parasite, and also to Nosema.
... if they spend enough time on Duolingo ...
If you’re planning on moving your bee colony to Thailand, have them brush up on their cha-cha-cha. But sadly, the hallowed bee dances in old Bavarian halls have basically no function.
I remember friends at school who displayed a similar ability before they were about to enter an exam hall.
... or a Mexican wave.
Duh.
Here’s a ridiculous diagram of this [delete 'ridiculous'; it's a fine diagram]"