site banner

There’s no psychopathology.

I’d like to start with a few disclaimers. This is not an anti-psychiatry post. This is also not the place to ask or receive advice about your mental health, or what nowadays is called “mental health”.

For some time now I’ve been feeling like I live in a different world than most people I know. It has come to a point where I have to face an awkward alternative: Either most people I know are wrong (including learned men and experts) or I am insane. As I don’t believe I have lost my sanity, and as I believe that I have very strong arguments to hold my ideas against all reasonable counterarguments; I think it’s about time I sit down and share my ideas more or less publicly. This is one of such ideas. What follows is the summary of my academic studies, my professional experience working in the field of mental health, my own personal thoughts, and the influence of several authors, chiefly Georges Canguilhem and Jacques Lacan.

The APA defines psychopathology as “the scientific study of mental disorders, including their theoretical underpinnings, etiology, progression, symptomatology, diagnosis, and treatment”. It is a jurisdiction of medicine, although that does not exclude other disciplines from delving into it as well. It is intrinsically linked to psychiatry, to the point where one cannot exist without the other. But psychiatry itself is a rather contradictory branch of medicine, because while every other specialization of medicine has built its own object of study by exploring a specific organ or function from the body, psychiatry exists only by virtue of that which it ignores. In its origins, psychiatry was born to deal with what has been classically classified as insanity, those people described by Descartes who believed they were made of glass or who fancied themselves to be pitches of water. These outlandish delusions have always caused turmoil in society because nobody really knows where they come from, what they mean and, most importantly, what to do with them. Insane people clearly need help but they do not want it, or what help they are willing to receive it’s impossible for other people to give. They break the law but they are not criminals, or at least they are bening. They behave like savages but are human beings and deserve to be treated as such.

Now enter the Enlightenment: Lady Reason triumphs all over the Western world, everything now has or will have a place and an explanation in the encyclopedia of universal knowledge. And what we understand we control. There are now a bunch of physicians who have little evidence but little doubt that these people are sick and that it is their task to heal them. And that they’ll try with all their available resources, but with little success. So while neurology developed from the study of the brain, cardiology from that of the heart and so on, psychiatry was born out of sheer embarrassment. It is the branch of medicine that studies mental disorders. However, being a part of modern scientific medicine, it cannot but assert that mental disorders can be explained by studying the body, the contradiction being that the day psychiatry discovers the bodily cause of mental disorders will be the day that it ceases to exist as a specialization of medicine, for said cause would fall under the jurisdiction of another specialization: If it’s in the brain then it would be neurology, if it’s in the genes it would be medical genetics, and if we were to discover a new organ in the body then a new specialization will be born to study it, leaving psychiatry in the past.

Therefore, psychiatry exists only because we do not know what mental disorders are. In fact, we don’t even know if the mind is real or not, much less whether it can get sick. What do we actually know then? We know that 1. there are people who need help, and 2. that there are means to help them. So it becomes a matter of administering a scarce resource. This is what psychopathology really is: It is not a science of mental pathology, it is the art of distributing psychiatric drugs and psychological treatments.

There used to be psychopathology. Classic psychiatrists wrote impressive treaties on the subject, with thousands of pages explaining in detail and classifying the behavior of their patients. The mountains really were in labour, alas, only a mouse was born: No progress was made regarding the causes, and most importantly the treatment of such behaviors. This last problem was drastically improved by the invention of psychopharmacology. Suddenly psychiatrists had a powerful tool to treat the symptoms of insanity, so even though they weren’t any close to understanding these symptoms, they changed their ideas on the subject to reflect the influence of psychiatric drugs. These influences can be accurately gauged by the changes on the DSM. The first DSMs included theories about the origin and nature of mental disorders, the last DSMs only mention the clinical symptoms necessary to prescribe a treatment. When a patient is diagnosed with depression the only relevant information that is learned is that said patient will start a treatment for depression.

So are mental disorders real? Of course they are. Whether they are mental or disorders, that’s another question. They are real because they are a set of behaviors that have been observed to occur together: Feelings of sadness, self-harming ideas or behaviors, inability to feel pleasure, these are all things that are real, observable, measurable, and treatable. But are these symptoms a mental problem? Are they a medical problem, or a problem at all? This is highly debatable, and in any case, not a solid foundation for a science.

If a person feels sad all the time, it is only natural for them to think that this life is not worth living. But the opposite is also true: If a person is convinced that there is nothing good in this world, then they will feel sad and hopeless all the time. So what comes first? Should we treat the sadness or the thoughts? And what if the person likes to feel sad, if they don’t want any help? Should we force them? And to make matters worse, it turns out that both psychiatric drugs and psychotherapy are effective*. And this is only to talk about those treatments that have empiric evidence to back them up and are approved by psychiatry, because, under the right circumstances, literally everything can be therapeutic: There’s horse therapy, art therapy, music therapy, dog therapy, video-game therapy, you name it.

There are some who believe in the demon deceptor, a person, or a group of people, who control our reality and make lies pass for truth, usually with malicious intent. These people believe that the pharmaceutical industry has created mental disorders only to sell drugs, and that psychologists and psychiatrists are their accomplices. For my part, I think it is overly optimistic to believe that someone has such a degree of control over the situation as to make it bend to their will. I believe that people are just confused, and with good reason, because being human is quite a bizarre experience. There are of course those who profit from the confusion of their fellow man, and prey on their ignorance. But even evil has its limits, and nobody can summon such perfect wickedness that no good may come of it. The truth is that for all the confusion that our idea of psychopathology entails, the treatment and the care for people with mental disorders has progressed a great deal in the last decades.

On the other hand there are the encyclopedists, who will argue that the fact that we haven’t discovered the bodily sources of mental disorders does not mean that we won’t succeed in the future. We have certainly made discoveries in this direction: Not only do we know now that it is impossible to be sad or mad without a brain, but we also know what specific brain part or substance is required. But even after all the advances in neurology, still no neurologic exam is indicated for the diagnoses of mental disorders, and for good reason. Because ultimately, what decides if someone has a mental disorder or not are arbitrary criteria. The fact that homosexuality is no longer a mental illness is only because of the fact that society has shifted its values towards the acceptance of diverse sexual orientations, were it not for that fact we would speak about the “homosexual brain” just as we know speak about “the depressed brain”. We could also speak about “the carpenter brain” or “the the writer’s brain”, and treat all of those conditions as illnesses.

In conclusion, I believe that contemporary psychopathology is a case of finding a hammer and suddenly realizing we are surrounded by nails. If something can be treated as an illness it will be treated as an illness, because that is l’esprit de l’époque. Classifying something as an illness, assigning it a part of the brain, and prescribing it a drug as treatment makes it real and important, so politicians, scientists, and the general public are aware of its existence and direct resources its way. This is why everyday we “discover” that there are more things linked to mental health: Poor housing, poor nourishment, the weather, sexual orientation, racial discrimination, political ideologies… and as there is no psychopathology there’s no limit to psychic pathologies. There’s a drug for everything or a therapy for everything. It’s no coincidence that we now have the most effective treatments in history and the highest rate of accessibility to mental health services ever, but the rates of mental disorders are soaring well. And despite all the advances in psychotherapy and psychopharmacology, no breakthroughs have been made in psychopathology.

I’m convinced that in the future people will look at our ideas on psychopathology as we now look at humorism.

Sources:

APA Definition of Psychopathology: https://dictionary.apa.org/psychopathology

*Psychotherapy just as effective as pharmacotherapy: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5244449/

3
Jump in the discussion.

No email address required.

And what if the person likes to feel sad, if they don’t want any help? Should we force them?

Yes, absolutely! https://youtube.com/watch?v=zQGN0UwuJxw&t=412

Literally me.

It has come to a point where I have to face an awkward alternative: Either most people I know are wrong (including learned men and experts) or I am insane.

I'm going to treat this as a request for an explanation of what is going on in the heads of the learned men and experts.

They are trapped in a language game, made toxic by linguistic poverty. My explanation will involve coining two neologisms and deploying them in a two stage explanation.

semi-grift Contrasts with the existing word grift. Grifters get prosecuted. But some grift-like enterprises do more good than harm. Even when individual practitioners do more harm than good, they still don't get prosecuted, because society doesn't want the chilling effect on the others, doing more good than harm and producing (in the case of psychiatry) a very substantial net benefit. Call those pro-social grifts semi-grifts.

turbo-real Are bricks real? Drop one on your foot and the pain will tell you: Yes, absolutely real! Are centers of gravity real? Put your brick on the table with the center of gravity over the edge. When it falls off and lands on your foot, the pain will tell you, err, what exactly? The center of gravity is too abstract to be absolutely real, yet the pain insists that centers of gravity are of such practical importance that they could only be one degree less real: turbo-real.

stage one The vocabulary of the elite includes semi-grift and turbo-real. You ask a member of the elite whether psychopathology exists. You persist. Eventually, when you are both naked in the hot tub, he decides that you are not wearing a wire, and the noise of the bubbles and water jets will hide the conversation from prying microphones. He levels with you: psychiatry is a semi-grift, psychopathology is not turbo-real.

Why the precautions? He explains that if he is overhead saying "psychopathology is not turbo-real" the normie eaves dropper will look up turbo-real in the elite-to-normie dictionary and find that it means that psychopathology does not exist. Then the normie eaves dropper will translate the other way. Since the normie term "exist" smushes together both turbo-real and semi-grift, translating back to elite speak from exist also hits semi-grift. Psychopathology is not even a semi-grift, it must be a grift, it must be banned and grifters prosecuted. Your elite informant is terrified of the disaster that would be caused and that is why he is so cagey. end of stage one

Now we must confront the impossibility the conversation above. semi-grift and turbo-real are my neologisms. No-one can give you the answer above because they don't have those words.

stage two In reality the elite suffer the same linguistic poverty as the normies. They have only one word, exist, which they decorate with elaborate qualifiers, hoping to distinguish strong forms of existence, such as being turbo-real, from weak-forms of existence, such as being merely a semi-grift. They know that their single word conceals a real distinction, that they grasp towards. They know that the stakes are high and being misunderstood by normies would be a disaster. Fearing that their struggle to be understood correctly might fail, they clam up.

This is where you end up bashing your head against a brick wall. The learned men and experts are surely clever enough to see that psychopathology is not turbo-real (though you use the word exist). But they will not come out and say it. Sometimes they fob you off by saying "Of course psychopathology exists" but that only adds to your confusion, because you can read people well enough to tell that they don't really mean it (or don't exactly mean it, or don't fully mean it, or something; there is definitely something hidden and wrong). The experts hear your questions. They smell disaster in the air. They don't have the words to articulate the actual difficulty, so they clam up, leaving you doubting your sanity. end of stage two

Basically a mix of abstraction, reification, and equivocation, right?

I don't like this explanation because it supposes a high amount of consciousness on the "elite"'s side, and I don't think that the experts really know what they are doing. In fact, I believe nobody knows, and that's why we have experts in the first place. You can be very good at something or very knowledageable in a specific subject, but what does that really mean? Isn't it all vain in the end? The truth is that people do things because there's nothing better to do, and sitting under a tree to medidate for 75 years isn't really an option for most of us.

For me the best explanation is that we are all very confused and very prone to deceive ourselves, so stumbling upon the correct answer is a matter of luck. And any answer that does not take this confusion into account is cheap and probably false, but it's also easier to ellaborate and disseminate, so opinions held by the masses are usually truisms or gross misunderstandings. It's also quite common that people are intelligent for something and dumb for something else, it happens all the time. I've seen very intelligent people do the following: They like X. One day, X does Y and they remain silent. The next day X's opponent does Y, and they start loudly criticizing X's opponen for Y, without noticing the contradiction. We cannot understimate people's ability to deceive themselves.

Rolls up sleeves

Aight, I'm here to pitch. I might not be a psychiatrist, yet, but it's that or die trying so I'll try and defend the much maligned field.

On an object level. You're mostly right. Psychiatry is largely pre-paradigmatic. We only have very vague ideas about the underlying underpinnings of many of the most common diseases.

We can find pretty reliable correlates in some cases, such as subtle differences in neuronal wiring or the activity of larger regions that associate with mental illnesses like depression, autism, OCD and so on. Unfortunately, much like the discovery of Tau proteins in Alzheimers, just because we have a marker for the illness does not mean that reversing the marker will reliably reverse the disease. All the drugs that claimed to reduce neurofibrillary plaques did jack-shit for actually curing Alzheimers.

However, being a part of modern scientific medicine, it cannot but assert that mental disorders can be explained by studying the body, the contradiction being that the day psychiatry discovers the bodily cause of mental disorders will be the day that it ceases to exist as a specialization of medicine, for said cause would fall under the jurisdiction of another specialization: If it’s in the brain then it would be neurology, if it’s in the genes it would be medical genetics, and if we were to discover a new organ in the body then a new specialization will be born to study it, leaving psychiatry in the past.

Oh boy.

For example, autism

Now, we’re going to get a little technical. Grey matter ripples into peaks and troughs called gyri and sulci, respectively. According to researchers from San Diego State University, these deep folds and wrinkles may develop differently in ASD. Specifically, in autistic brains there is significantly more folding in the left parietal and temporal lobes as well as in the right frontal and temporal regions.

“These alterations are often correlated with modifications in neuronal network connectivity,” Culotta says. “In fact, it has been proposed that strongly connected cortical regions are pulled together during development, with gyri forming in between. In the autistic brain, the brain reduced connectivity, known as hypoconnectivity, allows weakly connected regions to drift apart, with sulci forming between them.” Research has shown the deeper theses sulcal pits are, the more language production is affected.

Yay! We have a neurological, albeit still rough, understanding of what's going on here.

And yet, pray tell what a neurologist is going to do about it? Take an autistic child to a neurologist or neurosurgeon and ask for them to be cured. I'm sure they're too kind (for all their ego), to laugh you out of the room, but all they can offer is pity and a referral to a shrink or SALT.

Similarly, depression is primarily a disease of the brain (leaving aside conjectures about the gut brain axis, general inflammation and so on). Can a neurologist do anything about it?

Well, nothing a psychiatrist doesn't already do. In other words, therapy, meds and more aggressive interventions like electroconvulsive therapy (and ketamine, LSD and so on).

What have you achieved, barring a rebranding?

Even as we begin to understand some of the underlying processes of pathophysiology in depression, such as why rebooting the brain with induced seizures works when meds fail, it does not change the fact that ECT works beyond reasonable doubt.

In a world where we suddenly discovered, with perfect accuracy, the exact neurological underpinnings of mental illnesses (and didn't just cure them in the womb or immediately developed miraculous treatments), you know what would happen?

Specialization of labor. In other words "neurologists" doing the same shit as psychiatrists do today.

You're falling prey to semantics and fuzziness of definitions.

Geriatrics is a perfectly respectable specialization in medicine despite, last time I checked, there being no "old age organ" (well, maybe the thymus). Huh. Why hasn't their lunch been taken by all the cardiologists, oncologists, neurologists and other people with well-defined magisteria?

Specialization.

It is convenient.

It works.

On the other hand there are the encyclopedists, who will argue that the fact that we haven’t discovered the bodily sources of mental disorders does not mean that we won’t succeed in the future. We have certainly made discoveries in this direction: Not only do we know now that it is impossible to be sad or mad without a brain, but we also know what specific brain part or substance is required. But even after all the advances in neurology, still no neurologic exam is indicated for the diagnoses of mental disorders, and for good reason. Because ultimately, what decides if someone has a mental disorder or not are arbitrary criteria. The fact that homosexuality is no longer a mental illness is only because of the fact that society has shifted its values towards the acceptance of diverse sexual orientations, were it not for that fact we would speak about the “homosexual brain” just as we know speak about “the depressed brain”. We could also speak about “the carpenter brain” or “the the writer’s brain”, and treat all of those conditions as illnesses.

You poor man. You've been beaten to the punch by Dr. Scott Alexander Sisskind on several occasions, though I'm too lazy to link more than one, but which handily contains links to yet more ink spilled on the topic. Worry not, it happens to the best of us.

So are mental disorders real? Of course they are. Whether they are mental or disorders, that’s another question. They are real because they are a set of behaviors that have been observed to occur together: Feelings of sadness, self-harming ideas or behaviors, inability to feel pleasure, these are all things that are real, observable, measurable, and treatable. But are these symptoms a mental problem? Are they a medical problem, or a problem at all? This is highly debatable, and in any case, not a solid foundation for a science.

My answer is a resounding mu that brings all the local cattle to the yard.

None of this matters. Not that I really see any reason to call it "highly debatable".

You mistake medicine as practised as "science" whereas what doctors outside research settings do is closer to engineering.

I don't particular care that we don't know the exact cause of depression. I know of multiple batteries of tests that, with reasonable accuracy, tells me whether or not a given patient will benefit from counseling, medication, and other interventions, and which, when measured serially over time, tells me if it's working. And vice versa, if you take someone who doesn't have the markers of depression and feed them SSRIs, it doesn't make them happier. So what if it isn't actually a deficiency of serotonin that causes depression? The drugs, while less effective than desirable, are not useless.

I really don't see any reason to twist yourself into knots about whether depression is "mental", as opposed to what, neurological? Biochemical? All of the above in same capacity, depending on your appetite for abstraction?

"Medical"? Come on dude.

To draw an analogy, let's say you have a malfunctioning PC, or maybe a server rack in an AI datacenter. A vascular surgeon notices the water cooling is leaking and bills you $70000. A cardiologist checks the PSU. The neurologist makes sure the RAM is seated tightly or otherwise asks their neurosurgery buddies to bend some socket pins and reapply the thermal paste.

Whereas what psychiatrists are doing is both a combination of relatively simple things like giving drugs dusting the fans, as well as the equivalent of prompt-engineering an LLM. The brain, despite operating according to the same physical laws as an H100 loaded with GPT-4, is just as nigh impossible to understand from first principles, and some higher order discrepancies difficult to treat by looking at bare meat/metal. That does not mean we exist in utter epistemic helplessness. What we do works. It is not perfect. It is not complete. It is, however, not useless or a waste of time. What is is semantic arguments and forcing allied specialists to do jobs that they're no better equipped to do than we are, simply because that sounds neater and more scientific. Categories and doctors were both made for Man, and not the other way around.

Complex systems that aren't amenable to direct analysis at the elementary level are still amenable to modification and control. That is what psychiatrists do. If we were folded into neurology, then congratulations, all you would achieve is neurologists performing the same tasks a psychiatrist does, and likely end up creating a new subspeciality which is all but psychiatry in name (and we already have the opposite approach with some of the nerdier shrinks becoming neuropsychiatrists) .

This is the same line of thinking that would assume that, if one day we replaced the Standard Model with a Grand Unified Theory of Everything, then we no longer need those overly abstracted biologists, and those fussy chemists. Why, just model everything at the level of fundamental quantum mechanics? What do you mean the computer caught fire when I tried to diagnose ADHD using Feynman path integrals?

None of what you say is forbidden knowledge tacitly swept under the rug by psychiatrists, the witch doctors. The typical reaction, from anyone who has acquaintance with a textbook written after that fraud, Freud, would be the same long-suffering sigh as if you accosted an economist and told them that their models are flawed because they assume rational actors acting in enlightened self interest. They know that's not how it works. They're more than happy to fudge their idealized calculations or look for deeper trends. It's not an easy task, and in many ways, the interaction of many minds operating under relatively well-defined incentives is easier to model than a single one which is a fractal, sweltering, wet and buzzing ball of noise on the verge of criticality .

It's all rather moot at that point.

I appreciate your reply quite a bit. I’d like to ask you, in earnest, if psychiatry is legitimate, why is the problem only getting worse with increasing application? Has “mental health” ever been worse than now? I don’t actually know the answer, but it sure doesn’t seem so.

It's a long story, but if I had to sum it up: Legitimate/well-intended awareness campaigns can cause mass psychogenic illness, social media fads and the fetishization of therapy culture doing the same, previously under-diagnosed conditions being actually noticed and diagnosed, the drug epidemic, and a reaction to social atomization and the fraying of familial bonds that contributed a great deal to people keeping their shit together in the past when they lived in objectively shit conditions.

It seems to me that it's both true that conditions like ADHD and autism that were once missed because of ignorance are now being diagnosed far earlier and there's a lot of fad chasing and social contagion. Do enough hard drugs and you'll break something up there too, and said drugs have never been more common or tacitly tolerated.

(The actual incidence of "real" autism has likely increased significantly since the previous century, potentially due to increasing parental age, but it's all very confused and I'm only conveying my best understanding of the situation in the West.)

Respectfully, you are missing the point. I agree with what you say, this is why I started by saying this is not an anti-psychiatry post. What I'm saying doesn't change anything in clinical practice. I never said psychiatry doesn't work, quite the contrary. I explicitly say that it exists because it works. I think you have the right mindset by thinking of it more as an engineering practice, you are actually agreeing with me there.

If you think this is just semantics then fair enough, but I do believe the way we speak shapes the way we understand things. This is what it's about, our concept of psychopathology. You mentioned geriatrics, but geriatrics is an epistemologically sound specialization of medicine. What comes first, the loss of muscle mass or aging? In this case we can clearly stablish an object of knowledge even if we can't do anything to stop aging. The same happens with dementia: What comes first, the loss of cognitive function or the proteinopathy? Note that we don't know what causes the proteinopathy, but we do know for sure that Alzheimer's is a brain disease. So I ask again, if depression is a brain disease, then what comes first? Until we can confidently answer that, we can't really say that depression is a "disease like any other". The fact that changing someone's brain is a treatment for depression does not mean anything, because we could also make a carpenter change their profession by messing with their brain, and as I said, both being depressed and being a carpenter change your brain in predictable and observable patterns.

Let's see an example: There's a person who went through a break up and a year after that they still can't get over it, they feel sad all the time, they don't enjoy anything, and they think they won't be able to live without the person they love. Of course this is a problem, of course this person needs help, and thankfully we have the means to thelp them, that's wonderful. But are they sick? Is this because of a problem in their brain? We don't know. Unless we do, on which case you'd have to illustrate. This is a question that nobody is asking because they are busy choosing who gets the zap, as we cannot give psychiatric treatment or psychotherapy to everyone in the world, even though most people culd benefit from it at some point in their lives. But I do believe it is a question worth asking, even if it were only out of curiosity for the human condition.

I never implied this is some sort of forbidden knowledge, and this is what baffles me the most. Even though it is evident that there is no such thing as psychopathology, most people act as if there were. I know this doesn't really change anything but I can't understand it, this is what moved me to make this post. I also never implied psychiatrists are witch-doctors, because even though psychiatry has epistemological issues it is still a legitimate branch of medicine because of the fact that it works. Note that I'm saying that psychopathology does not exist, not that psychiatry does not exist, because it seems as one can exist without the other. I would prescribe you a text by Georges Canguilhem called What is Psychology, it is about psychology but many of the issues he points out are also applicable to psychiatry.

Regarding the blog by Dr. Sisskind, I fail to see how his argument changes anything. He's a psychiatrist (I think?) talking about a working definition of mental illness that serves his profession. If there's anything specific you think is relevant then please point it out. But the article does talk about evolutionary psychology and oh boy do I have something to say about that. I will make a post about it soon, but for now let's just say that it tries to explain something we don't understand using something we understand even less. It is another of those things that I feel like everyone's playing a prank on me by believing it's legitimate. The good thing is I can rest assured knowing that I'll have psychiatrists to take care of me.

I'll end with a fun fact: Psychiatrists are called "shrinks" precisely because they were in love with Freud, or as Chesterton and you pronounce it, Fraud. This is of course a translation issue, and the anglicized version of Freud has long been proved to be a flunk.

This is what it's about, our concept of psychopathology. You mentioned geriatrics, but geriatrics is an epistemologically sound specialization of medicine.

I am confused by what standard geriatrics achieves legitimacy and psychiatry doesn't.

You seem to think that the mind is a nebulous entity, and thus if a disease can be attributed to the brain, it's the domain of neurology. All well and good, but diseases of aging is an even less clear-cut specialization.

At the very least, the causality of:

What comes first, the loss of muscle mass or aging? In this case we can clearly stablish an object of knowledge even if we can't do anything to stop aging.

Aging is a multi-causal process. Everything from telomerase degradation to oxidative stresses, super-exponential degradation in the surveillance capabilities of the immune system such that after a surprisingly tight threshold everything starts breaking down.

But each of the component aspects are within the domain of independent specialities. If you find this an acceptable state of affairs, then what's wrong with psychiatry?

All the patients have in common is that they're old. At that point, a domain specialist who works with the diseases predominantly found in the elderly becomes valuable. All the more when the emphasis switches from cure to comfort, as is often the case.

What comes first, the loss of cognitive function or the proteinopathy? Note that we don't know what causes the proteinopathy, but we do know for sure that Alzheimer's is a brain disease. So I ask again, if depression is a brain disease, then what comes first? Until we can confidently answer that, we can't really say that depression is a "disease like any othe

The question remains that whether comparing the onset of cognitive dysfunction or protein accumulation is even the right question to be asking. Hell, very recently there's been conjecture that it's a prion illness off incidents of what are attributed to horizontal spread, and neurosurgeons get Alzheimers at higher rates, potentially by aerosolized spread. Still a brain disease of course, but by god is not simple or the billions we've spent would have solved it.

My central thesis is that it is irrelevant how you class depression, since it's not clear to me what you mean by it being/not being a "disease like any other". Diseases are very heterogeneous, and attempting to do a neat taxonomy is asking for brain worms yourself.

Let's see an example: There's a person who went through a break up and a year after that they still can't get over it, they feel sad all the time, they don't enjoy anything, and they think they won't be able to live without the person they love. Of course this is a problem, of course this person needs help, and thankfully we have the means to thelp them, that's wonderful. But are they sick? Is this because of a problem in their brain? We don't know. Unless we do, on which case you'd have to illustrate.

I once again claim that is an irrelevant question. What does matter, both to the doctor and the patient, is whether any interventions are warranted.

Can we expect SSRIs to work? Are they worth trying with even the low success rates and effect sizes here, keeping in mind unpleasant side effects? Can we expect that the patient will recover with therapy alone?

These are questions that have relatively clear, if probabilistic answers.

The question of whether or not he's "sick" is just a dangling pointer which, occasionally, can be useful to resolve, especially when it comes to the purpose of insurance. To ascribe it much greater significance is pointless.

Imagine Lionel Messi, who would have been a dwarf (or very short) without HGH injections. Was he particularly happy about it? Was he sick? There are no clear cutoffs beyond convenience. In his case, any imaginary unhappiness was solved with a jab and superstardom, but he could well have been a depressed short dude really good at football and bitter about never making it.

At least in depression, we make an effort to look for proximate causes, if your mom dies, it is entirely reasonable to be sad for weeks or months. Not years on end. In the latter case, therapy and meds will likely help. Hence the label of depression slapped on you.

Is this because of a problem in their brain? We don't know

I think we do know. Electroconvulsive therapy works on the brain. It produces obvious results on the EEG and the patient's self-reported happiness, as I've linked before.

If it looks like a duck, quacks like a duck, and is amenable to birdshot, you probably have a duck.

As I've previously discussed, it is entirely possible there are contributing factors like the gut-brain axis and systemic inflammation, but the ultimate cause is in the brain, otherwise acting on the brain should not be expected to have such clear effect, especially when more indirect routes fail.

This is a question that nobody is asking because they are busy choosing who gets the zap, as we cannot give psychiatric treatment or psychotherapy to everyone in the world, even though most people culd benefit from it at some point in their lives. But I do believe it is a question worth asking, even if it were only out of curiosity for the human condition.

Once again, I say that pragmatism reigns. Is obesity a disease? That's certainly a debate for the ages. But if Ozempic was shown to be cost-effective, and a net return on investment, if insurance was arbitrarily forced to accept obesity as a "disease" and cover it, that would be the sensible thing to do. If it saved insurers on future costs, they'd spring for it themselves. Resource constraints are independent of whether or not it meets your standards of illness. I agree there isn't enough money to give everyone a shrink on call, but whether or not we should aim for that is a matter of pragmatic tradeoffs and not semantics.

I never implied this is some sort of forbidden knowledge, and this is what baffles me the most. Even though it is evident that there is no such thing as psychopathology, most people act as if there were. I know this doesn't really change anything but I can't understand it, this is what moved me to make this post. I also never implied psychiatrists are witch-doctors, because even though psychiatry has epistemological issues it is still a legitimate branch of medicine because of the fact that it works. Note that I'm saying that psychopathology does not exist, not that psychiatry does not exist, because it seems as one can exist without the other. I would prescribe you a text by Georges Canguilhem called What is Psychology, it is about psychology but many of the issues he points out are also applicable to psychiatry.

I apologise if I came across as implying you held those views, those were more overarching observations rather than specific critique.

I think psychopathology is a useful concept, even if it's a placeholder. To come back to LLMs, as is my habit, there are behaviors that are helpfully described with short labels, such as "mode collapse" and "hallucinations" even if we don't know precisely why they happen on the algorithmic level, even if the system is transparent to us. We can use these semantic labels, they are helpful in describing a phenomenon. The human brain is nowhere near as debuggable, so given its complexity its no surprise that the peccadillos of the software layer might be usefully tackled while (tentatively) ignoring the hardware, i.e therapy. To further pump the intuition here, it is not necessarily the case that a malfunction on the software layer is attributable to an error in the hardware, and a computer, be it silicon or meat, can well be functioning fine at the bottom but break above. Now, even this is physical. A buggy piece of software on an SSD is still physical, in that if you had enormous patience and an electron microscope, you could look at the system and find the faulty distribution of electrons in a NAND chip and fix it that way. That is self-evidently a terrible way to debug software, even if it is doable.

Similarly, while "mental" illnesses are obviously an abstraction on top of whatever the hell is going on in the brain, and then down to the quantum foam, being an abstraction is not a bad thing, especially when as I've shown, even understanding neurological underpinnings of some diseases hasn't helped so far.

We gave people ECT well before we discovered the EEG changes, because their psychopathological symptoms improved and that was good enough.

Regarding the blog by Dr. Sisskind, I fail to see how his argument changes anything. He's a psychiatrist (I think?) talking about a working definition of mental illness that serves his profession. If there's anything specific you think is relevant then please point it out. But the article does talk about evolutionary psychology and oh boy do I have something to say about that. I will make a post about it soon, but for now let's just say that it tries to explain something we don't understand using something we understand even less. It is another of those things that I feel like everyone's playing a prank on me by believing it's legitimate. The good thing is I can rest assured knowing that I'll have psychiatrists to take care of me.

I am aghast that someone has washed up on these shores and not heard of Scott. I suppose weirder things have happened haha.

I would strongly advice you read the entire article. It is enlightening, and if you have yet to enjoy the glory of his work, well I envy you for it.

In case it isn't obvious, I agree with plenty of the things you've said. The thing is, there are aspects where I do disagree, and in several cases, it's from a pragmatic viewpoint.

It seems clear to me that depression is primarily a disease of the brain, even if other things can contribute.

Both neurologists and psychiatrists grapple with dysfunctions of the brain, but at different levels of abstraction. That's the difference between an electrical engineer and programmer. Subsuming psychiatry, or declaring psychopathology useless achieves nothing, because all it does is offload the work to another branch, who will, because the incentives are so obvious, immediately create neurologists-who-are-psychiatrists-in-all-but-name. Similarly, a sufficiently talented (and insane) EE could create a large program, nay, any program by fiddling with physical bits, but there exist clear reasons not to do that, and benefits from having them be different specialities, even if they work on the same "hardware".

Further, arguing about semantics as to whether depression is a "disease" or a "mental" illness achieves nothing.

Is it a useful signifier that constrains expectations? In other words, if I tell you I am depressed, does that provide useful information? Yes. (For the record I am, and the first and second line psychiatric interventions have failed me, I am seriously considering ECT).

Does it give me more information about whether or not the tools at my disposal can help a patient? Yes. If you meet the requirements on one of the many depression indices, or the DSM criteria, then "medicine for depression" will likely help.

There are no more questions left to answer that matter. Perhaps if there's political valence, and people wish to wrangle insurance companies or governments to provide therapies of their choice, such as gender-affirming care (let's not go into that can of worms, I beg you). But even then that's just a matter of how the die fall and practicality. The wine sparkles regardless of whether you label it champagne.

I don't understand why you keep trying to defend psychiatry when nobody's attacking it. I guess you wrote your answer as you read the text because you literally quoted my saying that psychiatry is legitimate. Psychiatry may be quircky and not as epistemologically sound as other branches of medicine, but it is still medicine, still scientific, and still helpful. I would say though that psychopathology is a placeholder. I understand that you have a pragmatic point of view, and you really should. This doesn't really change anything for psychiatrists, and they don't need to worry about this, so I thank you for taking the time to engage in this discussion. But should we really be satisfied by having a placeholder instead of a psychopathology? I don't know if you agree that your belief that depression is a brain disease is just that, a belief, but I believe that's the case because the evidence I've seen is not conclusive. Yes, it's impossible to be depressed without having a brain and depression is something that exists and changes your brain, but here is when it's important to think what comes first. We can agree that treating people with depression should be our first and foremost concern regardless of existential questions, but why stop at that? Why not try to understand what's going on? This shouldn't change the attention and the care we give to depressed people, but it can help us think more acurately about the problem and who knows, maybe even come up with more effective solutions in the future.

Furthermore, I think that believing that psychopathology is something different than a placeholder opens room for all sorts of abuses. Psychiatrists are scientists and know the limitations of their discipline (or at least they should), but psychopathology is being used in all sorts of contexts where it has no business whatsoever, and this is in part because it is an epistemologically bankrrupt concept. Just as a currency undergoes inflation and looses value, concepts that are no rigorous enough are more likely to be overused. Psychologists are the worst offenders here, or maybe it's just my perception because I'm a psychologist myself and of course I know many people in my profession, so I'm surrounded by people who very loudily make all sorts of claims. The best psychologists I know are aware that psychopathology is a placeholder, but most don't or ignore that fact. And as psychologists feel insecure about the scientific status of their discipline they overrely on psychopathology to sell their services.

There is something very funny about the history of psychology, because as you must know computers where made with the specific objective to imitate human thought. But then in the 70's a bunch of psychologists saw computers and were astonished at how much they reasembled human thought, and came to the conclusion that the human mind works like a computer. I'm personally against the expression "Artificial Intelligence" because computers are neither intelligent nor dumb. They do what they are programed to do. An animal, for instance, can be intelligent or dumb because it is directly involved in the outcome of its decissions, and they can be wrong or right. Computers are never wrong, therefore they lack the ability to be implied in their decissions. So even if LLMs resemble human speech, we would be wrong to believe that speaking to an LLM is the same as speaking to a person. In that sense, just the fact that we can treat depression as a brain disease does not mean that it is a brain disease. This is only technically correct because it ignores the problem by fixing over it.

We can agree that treating people with depression should be our first and foremost concern regardless of existential questions, but why stop at that? Why not try to understand what's going on? This shouldn't change the attention and the care we give to depressed people, but it can help us think more acurately about the problem and who knows, maybe even come up with more effective solutions in the future.

Please don't interpret my fervent appeals for pragmatism as a lack of curiosity. If I was genuinely uninterested in such matters, I wouldn't even know the little tidbits of information I've sprinkled in! They're not in medical textbooks or my exam curriculum for sure.

Knowing more would be great. But do not expect that to necessarily mean that psychiatric treatment has firmer grounding. A lot of shit works and doesn't work and we don't know why. As with autism, even when we know why, we can't fix it (without more advanced gene therapy).

Psychiatrists are scientists and know the limitations of their discipline (or at least they should), but psychopathology is being used in all sorts of contexts where it has no business whatsoever, and this is in part because it is an epistemologically bankrrupt concept

Psychiatrists aren't scientists! Doctors, as a matter of course, are not scientists! Some of us do research and clinical studies. That isn't our core responsibility, and most doctors you see have no papers to their name.

We are engineers. We try and fix things, and if a tool works, it works. That does not stop us from seeking better tools.

I looked at the Wikipedia article again on psychopathology, and as far as I can tell, it is an entirely benign subject and I am fundamentally confused by accusations of it lacking epistemically bankrupt concept.

Psychopathology is the study of abnormal cognition, behaviour, and experiences which differs according to social norms and rests upon a number of constructs that are deemed to be the social norm at any particular era.

Biological psychopathology is the study of the biological etiology of abnormal cognitions, behaviour and experiences. Child psychopathology is a specialisation applied to children and adolescents. Animal psychopathology is a specialisation applied to non-human animals. This concept is linked to the philosophical ideas first outlined by Galton (1869) and is linked to the appliance of eugenical ideations around what constitutes the human.

Later:

Psychopathology can be broadly separated into descriptive and explanatory. Descriptive psychopathology involves categorising, defining and understanding symptoms as reported by people and observed through their behaviour which are then assessed according to a social norm. Explanatory psychopathology looks to find explanations for certain kinds of symptoms according to theoretical models such as psychodynamics, cognitive behavioural therapy or through understanding how they have been constructed by drawing upon Constructivist Grounded Theory (Charmaz, 2016) or Interpretative Phenomenological Analysis (Smith, Flowers & Larkin, 2013).[7]

CBT is slightly better than the alternatives. I am exceedingly dubious that what it claims are the underlying mechanisms are what's actually going on, but it still works, and beats placebo and (barely but significantly) the alternatives.

I have done a deep dive on the topic myself, but I'd have to dig very deep into my profile to find it.

But even then, the existence of flawed models (which still do useful things) is no more a scathing critique than someone claiming that the Standard Model being unable to explain the overwhelming majority of the matter or energy in the universe makes Physics as a whole illegitimate. We know it's flawed. It's still useful.

There is something very funny about the history of psychology, because as you must know computers where made with the specific objective to imitate human thought. But then in the 70's a bunch of psychologists saw computers and were astonished at how much they reasembled human thought, and came to the conclusion that the human mind works like a computer. I'm personally against the expression "Artificial Intelligence" because computers are neither intelligent nor dumb. They do what they are programed to do. An animal, for instance, can be intelligent or dumb because it is directly involved in the outcome of its decissions, and they can be wrong or right. Computers are never wrong, therefore they lack the ability to be implied in their decissions. So even if LLMs resemble human speech, we would be wrong to believe that speaking to an LLM is the same as speaking to a person. In that sense, just the fact that we can treat depression as a brain disease does not mean that it is a brain disease. This is only technically correct because it ignores the problem by fixing over it.

The human brain is a computer. It just happens to not adhere to the Von Neumann architecture as most electronic ones do, but it is possible to simulate a single biological neuron with ~1000 artificial neurons in the ML sense..

Further, the human brain is bound by physics. Evidence otherwise is sorely lacking. We can simulate physics very well, at least if you don't want to use QM on macroscopic structures at typical temperatures, but only because that is a computationally difficult thing to do, not because it is fundamentally impossible to model.

Humans do what we were programmed to do. We just had a Blind Idiot God as a programmer, who had to bootstrap a VERY complex computer from a surprisingly small amount of code (DNA and epigenetics).

"Evolution, please grant me intelligence."

"To get more bitches and gather more berries?"

"Yes"

Invents condoms and ozempic like a boss 😎

You mistake the difficulty in unpacking the blackbox of human cognition as proof that it can't be unpacked. That is a grave error indeed.

Well, I'm at a bit of a loss here. What do you think engineering is if not the application of natural sciences? It's not the fairy-loving-godmother that engineers things. Claude Bernard would vivisect you for saying that doctors are not scientists, and then Kraepelin and Jaspers would electroshock some sense into your computer for saying psychiatrists are not scientists. Psychopathology could be a bening illusion, but the fact that believing in something that does not exist doesn't hurt anyone is no argument to hold that belief, especially when we can be just as effective without it as we are now.

For the rest, last time I checked the preferred psychotherapies were third wave behavioral therapies like Behavioral Activation and Acceptance and Commitment Therapy. Even if in practice they apply cognitive techniques, they ultimately are followers of Skinner and therefore assert that neither the mind nor cognitive mechanisms exist. They also sustain that depression is not a brain problem but a behavioral one, and many (like Marino Perez Alvarez) go as far as to question the relevance of psychopathology. I don't particularly adhere to this school of thought but I do agree with their conclusion regarding psychopathology.

Lastly, I don't assert the impossibility of unpacking the blackbox of human cognition. What I said is that we are yet to find definitive proof that it is possible, so it's not time to make claims about what humans are programmed to do or not to do quite yet. I would also say, paraphrasing an scholia by Nicolás Gómez Dávila, that if the universe were so artless as to be comprehensible to the human brain, then it would be immeasurably and unbearably boring, and we would have legitimate reason to feel disappointed. It's hilarious that everytime humankind creates some wacky artifice it believes it holds the key to understanding the universe. It happened with fire, writing, mechanical watches, and now computers, and so shall it be per saecula saeculorum. I guess there are computers everywhere for those with the eyes to see.

You mistake the difficulty in unpacking the blackbox of human cognition as evidence that it can't be unpacked. That is a grave error indeed.

Can you at least agree to the following:

  • that "unpacking the black box of human cognition" would involve the practical ability to have granular, read/write access to an actual human mind.

  • That no read/write access to a human mind has ever been demonstrated, nor has any meaningful progress toward such a capability ever been demonstrated.

  • That many people have previously claimed to be capable of demonstrating such access, or else of generating the capability to demonstrate such access, that their claims have been taken seriously, been tested rigorously, and have uniformly failed those tests.

  • That current iterations of the claim, such as yours here, no longer make straightforwardly testable predictions of the sort that were common from prominent scientists and "scientists" over the last century.

  • That the actual engineering we do with humans in fields like teaching, law and order, political organization and so on, all operate as though the self is not bound by physics in the way you believe it must be. That is to say, when a machine does something wrong, we go for the person who programmed it, but when a person does something wrong, we punish them directly. When we try to shape humans, we do so with techniques working from the assumption that the individual is autonomous and possessed of their own free will in all practical senses of the term.

I'm going to try to restate what I see as your position, before responding to it:

With regard to "read/write access", it appears that you don't mean it in the basic sense of "Do things that inform you of the content"/"Do things that change the content", but rather you specifically mean "outside of the normal IO channels". This is because free will is the big thing here.

Because I have free will, nothing you can do through my normal IO can control me. You can present evidence, and I'm free to veto the idea that it's even evidence. You can listen to what I choose to say -- or choose to think at your implant -- but you can't keep me from lying and you can't detect when I am. This fundamentally changes things because it means you cannot neglect my will; I am in control of how things pass into/out of my mind, and until you can go around my normal IO channels you need my buy in unlike with ships and planes who don't get a say in things. As a result, the normal paradigm of engineering ain't gonna work.

For "read access" to change things here, you would have to be able to not just read my surface level outputs but also the deep generating beliefs with reasonable resolution -- at least to the degree that "lie detection" can be done reliably. For "write access" to change things you would have to be able to write my conclusions not just impressions.

And reliable lie detection doesn't exist. It's impossible to "hack" into someone's mind in a way that bypasses the individuals say on things, and do things like "making a Christian into an atheist" or "implant a memory". Been tried, failed.

Is this essentially correct, or am I missing a key distinction here?

Because it looks to be like you're noticing that there's almost always a little white in a grayscale world and that attempts to do "pure black" aren't super successful, and then making the mistake of declaring everything to be "white" because it's "not [completely] black".

There's a lot of gray area out there, and some of it quite dark.

Is this essentially correct, or am I missing a key distinction here?

You nailed it. And specifically this part here:

This fundamentally changes things because it means you cannot neglect my will; I am in control of how things pass into/out of my mind, and until you can go around my normal IO channels you need my buy in unlike with ships and planes who don't get a say in things. As a result, the normal paradigm of engineering ain't gonna work.

...And further, that this view is supported by an overwhelming amount of evidence from every facet of human behavior, and every claim to the contrary is either unfalsifiable or has been falsified, yet people continue to insist otherwise, in a way identical to Sagan's invisible dragon. This isn't because they're stupid, it's because Sagan's invisible dragon is describing something irreducible about how humans reason. Reasoning is not simply doing math on accumulated evidence. The evidence is weighed and assessed in reference to axioms, and those axioms are chosen. You can choose to uncritically accept one provided to you by others, or you can choose to look at an arbitrary amount of arbitrarily-selected evidence until you arbitrarily decide that no more evidence is needed and a conclusion can be drawn, or you can take certain positions as self-evident and then prioritize the evidence that is compatible with them.

That last option is how people end up believing in Determinism, despite zero direct evidence in favor of determinism and a lot of evidence against it: they've adopted Materialism as an axiom, and Materialism requires Determinism. Any evidence against determinism is likewise evidence against Materialism, but because Materialism is an axiom, evidence against it is simply deprioritized and discarded. This is not objectionable in any way, and it is the only method of reason available to us. The problem comes from people ignoring the actual operation, and substituting it for some fantasy about reason as deterministic fact-math, as though their choices were not choices, but predetermined outcomes, and anyone who doesn't choose the same axioms is simply not reasoning properly.

There's a lot of gray area out there, and some of it quite dark.

I'd be interested on the grey you see. Torture regimes observably fail. Totalitarianism observably fails. Power slips through the fingers, despite all efforts to the contrary. People have been trying to reduce humanity to an engineering discipline for three hundred years running, and they've failed every time. Again, that's not conclusive proof that they'll continue to fail indefinitely, but looking at the historical record, and accounting for my understanding of technology that actually exists, I like my odds.

Saying "torture regimes fail" is like saying "cars fail". Of course they do; entropy is a bitch. But cars also work for a while before breaking down. It's neither the case that "Torture regimes never fail" nor that "Torture never accomplishes anything for the torturer". It's a question of "to what extent", and "in which circumstances?".

The difficulty of "engineering people" doesn't require determinism to be false, just that we have imperfect knowledge of what the determinants are. You'll have a hard time getting into my safe, despite the combination lock being entirely deterministic. If you were to have a sufficiently good model of the internals, you'd know just what to do in order to get the desired response 100% of the time. If you have a partial model, you only get partial results. It's just a matter of entropy.

Similarly, one's ability to persuade a person depends strongly on their ability to predict what kinds of things this person would veto as "not evidence" and what they would accept. Even if we assume human beings are 100% entirely deterministic, in order to get 100% results we need to have a complete model of the deterministic algorithm which changes by the moment as new experiences accumulate. We don't have to posit that a human mind is fundamentally non-deterministic in order to recognize that perfect determination is going to be an infeasible practical problem -- hence the "humans need to be treated like people" abstraction.

But what if we don't care about perfect 100% results? What if we don't limit ourselves to zero chance of failure, zero limit to the reach of control, zero limit to the duration of control?

Things get a lot more feasible. Now we don't have to contain a 100% faithful and ever changing model of the person we're attempting to "control" -- or perhaps more fittingly "manipulate". We just need to create a situation where we can reduce the entropy enough that we can get the results we're looking for before the entropy compounds and bursts through the seams.

And sure enough, manipulation works. Not well enough to get you a stable and fulfilling marriage into old age, but people do get manipulated successfully enough that it harms them and benefits their manipulators -- in the short term, at least. Serial killer Ed Kemper used to look at his watch and mutter something about not knowing if he had time to pick up a hitch hiker as what PUA would call a "false time constraint". Because the interaction of "picking up a hitchhiker" is such a simple low entropy scenario it doesn't matter if he can fully predict everything because all he needed to do was find that one little regularity that allowed him to "social engineer" some victims into his car.

A much more extreme version of this "funnel people into low entropy and take advantage of superior knowledge of the terrain" is hypnosis. Provided that the "subject" agrees to hypnosis and isn't creeped out and on guard, hypnotists can take advantage of a fairly low entropy set of possible responses to engineer ways to get people into states where their guards are predictably lowered even further, and then do stuff that bypasses the persons conscious will completely. Implanting fake memories is easy, and doesn't even require hypnosis. Implanting other ideas is doable too, as is prying out secrets that the person really does not want shared, and removing the person's ability to speak/move/remember basic things. The stuff that's possible with hypnosis is legit scary.

When you ask rhetorically "Can you make a Christian atheist?", my answer is "Provided they volunteer for hypnosis, yes, actually". I have run that exact experiment, and I forget my exact success rate but it was something like six attempts and five successes. The effect lasted about one to three months depending on the person, then they ended up reverting back to believing in God.

So is that "success" or "failure"? You could look at the bright side and note that it didn't last forever, or you could look at the dark side and notice that it worked remarkably reliably, for months without a shred of reinforcement, and with a very unsophisticated strategy and zero attempt made to make the effect robust.

It just comes down to what you're trying to justify. "Attempts to write the bottom line first and then engineer a way to manipulate people into doing what you want are unwise and ineffective in the long term and large scale", absolutely. "I know I saw that picture, because I remember it, and it's impossible to implant memories against a person's will", no.

As it applies to this conversation, it seems that the relevant question is "Can 'engineering' mindsets be used effectively to do things like help people with psychiatric conditions", I'd say "Yes, absolutely" -- but I'd also challenge your presupposition that "engineering" requires one to work around rather than with people's will. People's will can be predictable and controllable too, to an extent. Incentives shape wills, because people aren't dumb. If you show me a better way to get to work, I'll take it because it gets me what I want. Free will, sure. But also deterministic -- and determined by what gets me what I want. If you plug your fence into the electrical outlet, I won't touch it twice. Call it "operant conditioning"/"reprogramming when a person did something wrong", or call it "voluntarily deciding not to get shocked again". To-may-to, to-mah-to.

More comments
  1. To completely unpack it? Yes. The ideal would be to read individual molecules down to the limits of the Uncertainty Principle. Luckily even noisy signals like external EEGs provide useful data.

  2. https://youtube.com/watch?v=vpzXI1hlujw is a conclusive rebuttal. BCIs have been a thing for decades. We can reproduce imagery from dreams and even capture the mind's eye with surprising clarity, with non-invasive techniques to boot. I can write to your brain right now, just give me a scalpel, a needle, an electrode and medical indemnity documents.

  3. No. Or at least I don't know of any such people, I have no affiliation with them, and their failures do not impede progress in the field.

  4. No. If we emulate a human being, via brain scan or some high bandwidth side channel, and it works, then voila. That strikes me as about the same level of difficulty and temporal separation from today as a Victorian or early 20th century scientist theorizing about space flight and actual orbit. Or at least it would if not for AGI being imminent, which will likely solve the problem even if we didn't build it as a replica of the human brain (though inspiration was taken).

  5. What of it? Many human institutions are built on faulty foundations. "All Men are created equal"? They can pull the other one, even if it's a useful legal and social fiction. Things do not need to be true to do useful things, a monarchy backed by divine will has fuck-all going for it, it still manages to raise armies, collect taxes and build roads. You can scare a toddler with ghost stories and stop them wandering out to be eaten by a coyote.

Luckily even noisy signals like external EEGs provide useful data.

...People have claimed to have developed reactionless thrusters before. A test I've heard proposed is to hang the thruster and its power source from a pendulum, inside a sealed plastic bag, and then show it holding the pendulum off-center. In a similar vein, here's some proposals for similar tests of read/write access to the human mind: a working lie detector, love potion, mind-reader, or mind-control device would all be obvious demonstrations of the basic capability. Do you believe any of these exist, or that they will exist in, say, the next few years? If not, why not?

ttps://youtube.com/watch?v=vpzXI1hlujw is a conclusive rebuttal.

It certainly doesn't seem to be. I'm all for it, but this is reading output, not even input/output, and certainly not read/write.

We can reproduce imagery from dreams and even capture the mind's eye with surprising clarity, with non-invasive techniques to boot.

Could I get a cite on this? I would like to see some actual captures from dreams or the minds eye, because I'm pretty sure such things don't exist in the sense I understand the terms. I'm interested in being proved wrong, though.

I can write to your brain right now, just give me a scalpel, a needle, an electrode and medical indemnity documents.

You can damage my brain right now, or possibly jam it. You can't write on it in any meaningful fashion, as far as I know. Again, if you could, that would necessarily imply the present existence of mind reading and mind control, correct?

No. Or at least I don't know of any such people, I have no affiliation with them, and their failures do not impede progress in the field.

Marx? Freud? BF Skinner? Watson? As for affiliation with them, do you think the "God of the Gaps" is a reasonable criticism of Christianity? If so, is it just Christians who shouldn't collectively retreat to unfalsifiability, or are such collective retreats in the face of contrary evidence generally bad whoever is doing them?

No. If we emulate a human being, via brain scan or some high bandwidth side channel, and it works, then voila.

If. As in, in the future. As in, not in the present. You recognize that we not only can't emulate a human being, but we aren't anywhere close to being able to, right? That the capability you are reasoning from is, at the moment, entirely fictional?

What of it?

All direct evidence available to us shows that the human self has free will: we experience its operation every minute of every day, and can examine that operation in intimate detail. All engineering we do on humans operates on the assumption that free will exists, from education to law to navigating interpersonal relationships. Determinism makes no predictions that can currently be tested. Determinism's previous testable predictions have all been falsified. No engineering we currently do on humans operates according to deterministic principles. All these are facts, to the extent that "facts" can be said to exist.

The fact that all of the above can be so easily ignored can teach one important things about the operation of human reason, and particularly the prime role that free will takes in that operation. You can't be made to draw a conclusion by presented evidence, because the individual will has veto on what is even considered evidence. All beliefs are chosen by distinct acts of the will.

People have claimed to have developed reactionless thrusters before. A test I've heard proposed is to hang the thruster and its power source from a pendulum, inside a sealed plastic bag, and then show it holding the pendulum off-center. In a similar vein, here's some proposals for similar tests of read/write access to the human mind: a working lie detector, love potion, mind-reader, or mind-control device would all be obvious demonstrations of the basic capability. Do you believe any of these exist, or that they will exist in, say, the next few years? If not, why not?

A tailor jumping off the Eiffel tower with a large apparatus of mostly linen did not prevent the invention of working parachutes. The dozens of people jumping off cliffs to their deaths throughout the ages did not prevent heavier than air flight.

There are reasonably robust theoretical reasons to suppose that reactionless thrusters do not work. I wish to see the equivalent for BCIs.

It certainly doesn't seem to be. I'm all for it, but this is reading output, not even input/output, and certainly not read/write.

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

Foerster was the first to discover that electrical stimulation of the occipital cortex could be used to create visual percepts, phosphenes.[1] The first application of an implantable stimulator for vision restoration was developed by Drs. Brindley and Lewin in 1968.[2] This experiment demonstrated the viability of creating visual percepts using direct electrical stimulation, and it motivated the development of several other implantable devices for stimulation of the visual pathway, including retinal implants

(The eyes are also an extension of the brain from an anatomical and developmental standpoint, but you can write imagery to it directly, as the excerpt showing the effects of stimulation to the occipital cortex shows)

We can reliably produce different sensations and even cause muscular movements. The tech isn't at the point I can make you see HD pictures.

You can damage my brain right now, or possibly jam it. You can't write on it in any meaningful fashion, as far as I know. Again, if you could, that would necessarily imply the present existence of mind reading and mind control, correct?

Now, I can always adjust the voltage on the electrode, the indemnity documents are purely an insurance matter and not because I'm not licensed for neurosurgery. I can reliably induce plenty of somatic sensations in you and show you some really sweet colors, and once I get to the temporal lobe, I can promise meeting God Himself or your money back (and you don't even have to die, temporal lobe epilepsy or stimulation causes religious ecstasy).

We can read minds. See video linked. It literally picks up on his will to move his paralyzed arms and converts that to equivalent mouse movements. We can control minds. Just very crudely. The dude in the video says he intends to cosplay as Professor X, which is hilarious, and also not far from reality. If he can move a mouse cursor, he can move a robot, with his mind, at a distance. This has been done with Neuralink in monkeys, and with other BCIs, also in monkeys.

Mind reading and mind control exist. It's not a psychic phenomenon, it uses Bluetooth.

Marx? Freud? BF Skinner? Watson? As for affiliation with them, do you think the "God of the Gaps" is a reasonable criticism of Christianity? If so, is it just Christians who shouldn't collectively retreat to unfalsifiability, or are such collective retreats in the face of contrary evidence generally bad whoever is doing them?

Umm.. I have said some very uncomplimentary things about Marx and Freud. I believe I called the latter Fraud in a comment as recently as a day ago.

Skinner? I am neutral on him. The Skinner Box is an interesting idea, that's about all I can say from memory.

Watson? Presumably buddy of Crick? What did he do wrong? I have no idea.

I have seen plenty of Christians here happily engage in GOTG reasoning. You don't need to go back 50 years or more to find them. Yes, hiding in the dark and dank corners where you can just barely squeeze your eyes closed against the light of empirical inquiry is a shameful display.

But what's your point? I invite you to show me what I have in common with Marx, Freud, Watson and Crick. I am bemused that there exists a natural category the five of us share. I certainly denounce the former two much harder than I can recall any practising Christian here fighting against GOTG. If so, they have my thanks, I prefer an honest admission that they have an unshakable faith that I don't have to waste time debating rather than having to wait for the inexorable march of progress to squeeze them out of the gratings.

Could I get a cite on this? I would like to see some actual captures from dreams or the minds eye, because I'm pretty sure such things don't exist in the sense I understand the terms. I'm interested in being proved wrong, though.

This A.I. Used Brain Scans to Recreate Images People Saw

If. As in, in the future. As in, not in the present. You recognize that we not only can't emulate a human being, but we aren't anywhere close to being able to, right? That the capability you are reasoning from is, at the moment, entirely fictional?

The line between sufficiently hard scifi and a white paper is blurry indeed.

This is a terrible argument. You are likely using an electronic device that was "fictional" when first imagined to make it.

In the absence of AGI, I think it would take anywhere from 20-40 years to make a high fidelity emulation of a human brain. The bottlenecks are both available compute (the requirements to run a human brain on a computer are large, and the estimates vary OOMs). We also need better scanning tools, right now they're best suited for tiny slices at a time, and you can't do it while the subject is alive. Thankfully the latter is not a strict requirement, and shortcuts in compute and better algorithms probably exist. The record as it stands has long exceeded drosophila and roundworms, and the current SOTA is either an entire rat brain or 1% of a human brain.

If you disagree, please make it clear you're exceedingly confident in what I deem to be a very insane proposal that we will not have OOMs more raw compute in a few decades*, or that scanning techniques will not improve even modestly, given that new advances come out every year or two.

These are all difficult tasks. Hence the long time without AGI helping us. But nothing in my knowledge of physics, chemistry, biology or engineering rules it out, and the EU has enough confidence that it's possible that they spent a billion euros on the Whole Brain Emulation Project and are now working on its successor, EBRAINS. It's at about the same stage as the Human Genome Project was, and look where we are today. (Tbf, they did a terrible job at it. But it's a hard job nonetheless.)

*Short of civilizational collapse and mass extinction.

All direct evidence available to us shows that the human self has free will: we experience its operation every minute of every day, and can examine that operation in intimate detail. All engineering we do on humans operates on the assumption that free will exists, from education to law to navigating interpersonal relationships. Determinism makes no predictions that can currently be tested. Determinism's previous testable predictions have all been falsified. No engineering we currently do on humans operates according to deterministic principles. All these are facts, to the extent that "facts" can be said to exist.

The fact that all of the above can be so easily ignored can teach one important things about the operation of human reason, and particularly the prime role that free will takes in that operation. You can't be made to draw a conclusion by presented evidence, because the individual will has veto on what is even considered evidence. All beliefs are chosen by distinct acts of the will.

Please, not this topic again, I think we've been over this at least twice or thrice and I don't think we've made any progress. Certainly this is a starting point that would require rehashing what was a fruitless debate.

More comments

But then in the 70's a bunch of psychologists saw computers and were astonished at how much they reasembled human thought, and came to the conclusion that the human mind works like a computer. I'm personally against the expression "Artificial Intelligence" because computers are neither intelligent nor dumb.

All physical processes can be simulated by a computer. They are all computers. You seem to believe animal brains are special, but they aren't. An Intel CPU or an Nvidia GPU are both made of electrons and quarks, just like the neurons in your brain. Both are the same particles governed by the same laws of physics and just as computable as any other configuration of particles.

Of course animal brains are special, the fact that they can be simulated doesn't make them less so. The fact that you can take a picture of The Garden of Earthly Delights does not make it any less special. If you really want to be a nerd then no, nothing is made of electrons and quarks. In fact, nothing is made out of nothing and nothing exists at all, and if an atomic bomb were to explode right here beneath my chair nothing would change and everything would remain exactly where it is and where it has always been.

I believe this was written by an LLM.

FWIW I will still add my thoughts. The failing of psychiatry/psychology is in the categorisation of extreme personality differences as disorders. These are not actually disorders but phenotypes. The narcicistic personality, the sociopath, the psychopath, the high functioning autistic etc. are adaptive personality niches. The 'disorder' aspect comes in when the phenotype is so strong that the person cannot break the trait even when they are overwhelmingly incentivised to do so. But this applies just as strongly to say cowardice, which can be beneficial in certain circumstances and detrimental in others. The person who is constitutionally this way won't summon up the courage when it is in there interest but we don't call it a Cowardice Disorder. Likewise, the narcissist may destroy their whole family, but their self obsession may be very effective in pursuing goals that on net promote their reproductive fitness.

My personal issue is any portrayal of these people as victims. They exist as an unfortunate phenotype for society and have been classified as "disordered" due to their propensity to victimise others. I think it is correct to view it as a trait of the person they are responsible for suppressing, and if they fail to suppress it it is fully on them.

A new male lion taking over a pride will massacre all of the previous lion's cubs. Lions don't have personality disorders. They have a distinct phenotype which has been rewarded by natural selection. I think of antisocial human behaviours the same way. This leads to a harsher perspective on punishment.

I don't know if that is a praise of LLMs or an insult to my writing, but I do have to say that no LLM was used here. I did use a translator for some words and expressions because English is not my native language, and I'm aware that using "In conclusion" at the end is standard LLM style, I was aware of that while writing it, but I had to stop somewhere and I was too lazy to come up with something better. However, I think this is evidently not written by an LLM because I make some very bold and specific references, like Descartes' pitches of water. Of course an LLM could write that but you'd have to give it a very specific and long prompt, at which point it's just easier and faster to write it yourself.

Now, I don't think psychiatry fails at anything because its raison d'être is to treat people with mental problems, and it has a significant degree of success at that. It's certainly not perfect but it's way better than it was a hundred years ago and it keeps improving. For the rest, I belive that changing psychiatrists for other professions won't help us understand insanity any better, but it is true that modern society has a tendency to overanalize individuals and in doing so it somewhat justifies morally unaceptable behavior. Psychology is the worst offender here.

I don't know about you, but the ICD classification system we use in Czech Republic only classifies psychopathy as a disorder it causes problems to the person or others.

You can be a completely conscienceless type, as long no one's upset you aren't classified as a disorder.

For some time now I’ve been feeling like I live in a different world than most people I know. It has come to a point where I have to face an awkward alternative: Either most people I know are wrong (including learned men and experts) or I am insane. As I don’t believe I have lost my sanity, and as I believe that I have very strong arguments to hold my ideas against all reasonable counterarguments;

This is really less significant than you seem to be making it out to be, or beating yourself up over. The majority of people disagree with the majority of everyone else, including learned men and experts, on the most profound question possible. Religion.. No matter what religion you are, you're disagreeing with at least about 70% of people immediately due to not sharing a religion, and if you drill into the different denominations and sects that number grows larger. Add in ideologies akin to a religion and the number grows yet further. Are you an Atheist? Communist China has obligatory support for Communism, and unless you too believe in Xi Jinping Thought, then you still disagree with the majority of Atheists.

Anyway...

This is what psychopathology really is: It is not a science of mental pathology, it is the art of distributing psychiatric drugs and psychological treatments.

There used to be psychopathology. Classic psychiatrists wrote impressive treaties on the subject, with thousands of pages explaining in detail and classifying the behavior of their patients. The mountains really were in labour, alas, only a mouse was born: No progress was made regarding the causes, and most importantly the treatment of such behaviors.

A layman's explanation for how science works is that you make up a theory about how something works, you test it, try to disprove it, and then if you can't disprove it, you incorporate it into The Science. But this isn't quite true. It is a mistake to believe that science requires theoretical mechanisms to be explained before it can describe the world. Sometimes, things just are, and you can do science on that even if you don't understand why they are. Quantum physics makes incredibly good predictions about the behaviour of what it seeks to describe, yet it provides no theoretical explanation of why physics behaves that way, leading to several competing explanations.

A more apt comparison for medicine, however, would be general anaesthesia. A weird and miraculous phenomena where, upon breathing certain gasses or being injected with certain chemicals, your body enters a coma where it forgets everything that happens and stops experiencing pain. And when they are no longer administered, the coma reverses, and if the process is controlled correctly, you wake up unharmed! What function could this possibly serve? Why would the human body evolve to be capable of this feat in the absence of naturally occurring anaesthetics? But oh how useful it ended up being. It's a process we can predictably control and we even have a good idea about what sorts of molecules can cause it.

So what's the mechanism for anaesthesia? Why does it work? Nobody really knows. There's a few competing hypotheses but nothing to settle the question. But for the sake of using anaesthesia in millions of surgeries, that doesn't matter. Still works. We can still do it without having that explanation.

The same applies for mental illness. You don't necessarily have to understand it to treat it, though it could certainly help.

Firstly, I don't understand what you are trying to say. I mean, 99.9% of people feel sad at some point in their lives, but is that supposed to make me feel better when I'm sad? The fact that most people have unique opinions doesn't change how I feel, i.e. feelings don't care about your facts. I appreaciate you trying to help but I'd much rather keep being a snowflake and discuss with people on the internets.

Secondly yes, you don't have to understand something to change it, and ultimate causes are a theological issue, not scientific, as Thomas Aquinas divinely demonstrated. The point still stands: Psychopathology does not exist. We do not have a scientific knowledge of mental pathologies. This is an information you won't ever need in your life but people can stop pretending psychopathology exists any time now. On the other hand, if you are content with this and do not feel at least a bit curious about what insanity is and how it works then you are a bore and we don't have anything to speak about.

Lastly, we do know how anaesthesia works, lol. What was that all about? Of course there's no "mechanism", but there is such a thing as anaesthesiology.

If you're curious, the current best consensus is that most anaesthetics work by disrupting lipid membranes in neuronal cells.

This is, paradoxically, sometimes neurotoxic, especially in children, and sometimes neuroprotective, especially in the elderly.

I probably have a longer comment about it somewhere, but scrolling through several hundred on Reddit has failed to find it.

In conclusion, I believe that contemporary psychopathology is a case of finding a hammer and suddenly realizing we are surrounded by nails. If something can be treated as an illness it will be treated as an illness, because that is l’esprit de l’époque.

Does social dysfunction not matter at all in your paradigm? To a sufficiently disinterested observer, everything is culturally contingent, but the criterion of maintaining will to survival while not harming others without due cause seems trivially justified so long as we recognize the legitimacy of people, well, living in a society.

Lol, didn't see the disclaimer next to the user name.

Yes, but this does not change the fact that psychopathology does not exist. Of course social dysfunction is a relevant criteria to select the people who will receive psychiatric treatment, but would you say that criminals are all mentally ill? Certainly not.

We're in the "blood-sucking leeches" stage of psychiatry. The fact people are surprised whenever ancient chinese medicines or leeches work is funny, because of course they work. The recipes in Bald's Leechbook actually work -- were our ancestors were too dumb to notice patterns? The problem is they didn't know why. All knowledge reduces to probability, and the chance that an antidepressant works is around 10-20%. So our concept of depression is wrong. More accurately, it doesn't make sense. Like the "humors", it fails to explain anything, it's a non-sequitur.

There is a "real" psychiatry buried somewhere under this fake mess, or at least an improved set of models that works more consistently. The good psychs have put effort into this. But I am regularly surprised to see very smart psychs who are content with this system. It's they who should be leading the charge to improve, but they've accepted this paradigm of long-term patients with a 10+ year mental illness being failed by the system, they just shrug. "Tough luck!"

Yes, I agree. But I think that good psychiatry is close to what another person on this thread called "engineering". I don't think there will be much breakthroughs in psychopathology in the future, because as you say it is a non-sequitur. What we can do is give people with... mental problems? the best treatment we can, and in that sense thinking of insanity as an illness has dramatically improved conditions for them. For me the problem is that if we continue creating effective treatments without knowing what we are treating we open up Pandora's box, and who knows what will be considered a disease in the future. This is the characteristic trait of modern society: It falls prey to its own effectiveness.