Humans are artifact adoring artisans

[Coyote]

Well, he's forming his own sentences now, so it's a start.

Something shoulda' grown by now, considering the amount of fertilizer...

[SirNitram]

Oh get over it. No forum is going to be simultaneously an entertainment venue for flamewars and fundie-taunting as well a location for serious professional debate. Pointing out that SD.net is optimised for the former isn't 'belittling' any more than saying that comp.ai would be a bad place to argue Guri versus Data. That isn't a tactic to use against your opponents, that's just saying that Coberst would be better off somewhere else.

Why do you pretend that flamewars, and SWvST are all that's here? Are you really so blind that you've seen none of the debates about real world issues and the like, or is it too disruptive to your 'You're all extremists who only exist to flame' narrative to consider the fact you're full of it?

[Starglider]

Ok fine, I will take this idiot to task.

Cognitive science has two paradigms, symbol manipulation aka Artificial Intelligence and conceptual metaphor.

WTF is this? Did you somehow miss the entire subfield of connectionist AI, you know, the one that's been getting all the hype since the late 80s? You somehow missed the fact that every 1st year undergrad overview of AI, hell even the popsci ones, clearly divide the field into 'symbolic AI' and then all the other paradigms?

As for 'conceptual metaphor', firstly there's no way in hell you can bag up all the disparate work on metaphors (of which L&J is only one particular slant) into a 'paradigm', and even if you could that's a stupid and meaningless way to divide up the field.

You class work was evidently in the cognitive science of AI.

To what extent AI is a subset of Cogsci is still a subject of hot debate, not helped by the fact that many of the psychology people don't even acknowledge the concept of non-human (sapient) intelligences, much less 'cogsci' as generalised to them. This is a complete red herring though. Yes, the Turing test and the Chinese Room usually come up in an AI context, but people losing specific cognitive functions from specific localised injuries is the original bread and butter of high-level neuroanatomy. Many people have built their entire careers on working out the function of various areas based solely on painstakingly collecting patient reports.

Quotes from “Philosophy in the Flesh” by Lakoff and Johnson

But you're still not providing page numbers or even chapters, you know, the way even L&J manage to in their bibliography.

The ‘conceptual metaphor’ driven cognitive science considers the body and the mind to be one. The body and mind is a gestalt, the mind and body go together “like a horse and carriage”. Home determines to a large extent ‘who is Jane’; the body determines to a large extent ‘what is mind’.

Is this supposed to be in any way insightful? Yes dualism sucks, people are the sum of their memories and habits, you don't need to dress it up. Dennet did this vastly better in 'Consciousness Explained', hell, fucking Dr Who did this vastly better (scroll up to the quote in Stark's sig).

Western traditional philosophy

I've seen this a few times. Cogsci books along the explicit or implicit premise 'Eastern philosophy is so much more enlightened' invariably come off the rails. They tend to be massive rationalisation sessions for a personal preference. Even Hofstadter fell for this on rare occassions.

and the philosophy supporting AI are more like Jim than Jane insofar as the mind body relationship is concerned. For AI the body is a machine that functions with software that is used for mental operations. Like our PC the body is hardware that performs the operations the software dictates. As Jim considers home to be a place to eat and sleep so AI considers the body is a place for satisfying the need for a place for the mind to abide.

Screaming 'there is a metaphor here!' does not make it so. If there's something the poor ignorant AI people are missing, you'd better be clear about what it is, what it does and why we need it, or we will start sharpening the blade on Occam's Guillotine and checking your neck for fit.

We have in our Western philosophy a traditional theory of faculty psychology wherein our reasoning is a faculty completely separate from the body.

Dualism is discredited in the eyes of anyone with a brain. If you mean the possibility of non-situated intelligence, that's another relatively subtle argument. There isn't a simple one-liner right answer on situated intelligence (other than 'humans are situated intelligences, duh'), because it's fundamentally a question of engineering (in AI) and incremental paths (in evpsych). Based on your performance to date I guarantee that your position on that question will be easy to tear apart.

“Reason is seen as independent of perception and bodily movement.” It is this capacity of autonomous reason that makes us different in kind from all other animals.

Yes. People can do IQ tests. Dogs can't. How we got that capability is irrelevant to the fact that we can in fact do it. Deal.

I suspect that many fundamental aspects of philosophy and psychology are focused upon declaring, whenever possible, the separateness of our species from all other animals.

Your strawman notions are irrelevant. Certainly there was plenty of this about in times past, most of it religiously motivated. Current cogsci and even most philosophers at this point very much acknowledge that humans are a limited upgrade to the mammal baseline. But that upgrade was enough to give us total power over every other species on planet earth. Given that L&J focus on language of all things, the single most obvious distinguishing feature between our species and every other animal. I have no idea what you're complaining about.

This tradition of an autonomous reason began long before evolutionary theory

Yes, it's called abstraction, it was the foundation for all maths, science and ultimately progress.

and has held strongly since then without consideration, it seems to me, of the theories of Darwin and of biological science.

I note that the really stupid bits are outside the bolded bits. Note that L&J were writing in the late 90s when the whole evpsych revolution was only just becoming 'mainstream'. Nearly ten years later hardly anyone in cogsci hasn't given it serious 'consideration'. Sure there are still a few useless sociology, psychology and philosophy hardliners still aren't paying attention, but who cares about them?

Cognitive science has in the last three decades developed considerable empirical evidence supporting Darwin and not supporting the traditional theories of philosophy and psychology regarding the autonomy of reason. Cognitive science has focused a great deal of empirical science toward discovering the nature of the embodied mind.

Isn't that the sound of L&J refuting your own point?

The three major findings of cognitive science are:

'Major' is subjective but good luck finding anyone who agrees with you.

The mind is inherently embodied.

Blatantly obvious from the start. What took a bit of disappointment to drive home was that attempts to model abstract thought with ungrounded abstract symbols only goes so far. You can do a lot of useful things with that, but not build a general AI or a model of 'central cognition' in humans.

Thought is mostly unconscious.

Blatantly obvious from the start in the sense that humans don't know exactly how we do most of the things they do. Anyone with a passing familiarity with evolution realised that this is because reflective capabilities take up valuable wetware and processing time and the cost/benefit ratio doesn't make sense in selective terms. Humans already have much better reflection than all other animal species, that's what 'self-aware' means.

Abstract concepts are largely metaphorical.

Ok, this one is genuinely non-obvious. Philosophy wasted a lot of time with 'world of forms' and ungrounded concepts. L&Js stuff provides a big catalog of examples of how abstract concepts are grounded in humans. Unfortunately they don't provide much in the way of proposed mechanisms and (as far as I can recall) nothing predictive. They talk endlessly about what concepts are, but not how they work, which is what all the useful people do.

“These findings of cognitive science are profoundly disquieting [for traditional thinking] in two respects.

For religious nuts maybe.

First, they tell us that human reason is a form of animal reason, a reason inextricably tied to our bodies and the peculiarities of our brains.

Don't expect anyone here to be surprised. This is L&J desperately trying to be profound again, though that's somewhat excusable if they're targetting popsci readers. It's true that assorted human reflective delusions mean that people don't realise how broken and quirky they are (compared to a currently abstract, or at least impractical, ideal). But again, other authors (e.g. T&K, who to their credit L&J do cite in glowing terms in PITF) make this point much better and with less pretension.

It's a pretty special case of 'animal reason' though when it allows us to go to the moon, hunt and domesticate any other species we choose and seriously contemplate destroying the entire biosphere, while our closest relatives (in the primate kingdom) stare at us uncomprehendingly and maybe learn a few hundred words, for carefully taught geniuses.

Second, these results tell us that our bodies, brains, and interactions with our environment provide the mostly unconscious basis for our everyday metaphysics, that is, our sense of what is real.”

'Everyday metaphysics'? That sounds nice but it's non-descriptive.

All living creatures categorize.

Blatant hyperbole. Bacteria do not categorise. Earthworms do not categorise. AFAIK the simplest creature proven to learn categories is a pigeon.

All creatures, as a minimum, separate eat from no eat and friend from foe.

They have hardwired responses designed by evolution. That has nothing to do with humans doing explicit categorisation using our neural hardware. It is 'emergent' categorisation (much as I hate to use that buzzword) that humans perceive in the end result of a multi-millenia evolutionary process.

There are trillions of synaptic connections taking place in the least sophisticated of creatures

Nonsense. I note you're not quoting here. Adult humans have somewhere around 100 trillion synapses at any one time. Bacteria have zero synapses; in fact they're considerably smaller than a neuron. A cockroach has a few million. A sea slug has somewhere in the region of 20,000.

and this multiple synapses must be organized in some way to facilitate passage through a small number of interconnections and thus categorization takes place.

'And thus categorisation takes place'. Oh I love it. This is exactly the kind of bullshit you can't get away with in AI, because you have to explain yourself in enough detail for a compiler to understand. There are thousands of theories relating to categorisation, with various ideas for internal structure, formation, modification and neural basis ususally involving multiple layers of organisation. Those people at least put some work into it.

Great numbers of different synapses take place in an experience and these are subsumed in some fashion to provide the category eat or foe perhaps.

in some fashion - but you did not. I was trying to argue that Mike was being too harsh above, but morons like you are feeding him ammunition to write off the whole field, and smearing serious (if pompous and overreaching) researchers in the process with your personal misinterpretations.

Our categories are what we consider to be real in the world: tree, rock, animal…

Ever tried writing a complete ontology? It doesn't work that way, even with a somewhat more sophisticated notion of combinatorial categories than yours.

Our concepts are what we use to structure our reasoning about these categories.

Worthless without detail, and why didn't you just say 'categories are recursive via reflection', if that's what you meant?

Concepts are neural structures that are the fundamental means by which we reason about categories.

Point to a brain and show me the concepts. Go on, if they're distinct neural structures. Show me the neurons that form someone's concept of 'tree' or 'animal' or whatever, or at least tell me how to find them myself.

[Starglider]

Why do you pretend that flamewars, and SWvST are all that's here?

Strawman. I said that flamewars and making fun of morons are what this forum is particularly good at.

Are you really so blind that you've seen none of the debates about real world issues and the like,

Of course not. I've posted in a few. This forum is comparable to most others of similar size and quality in that respect. That sort of discussion is fine, it's entertaining (hopefully) and sometimes enlightening.

or is it too disruptive to your 'You're all extremists who only exist to flame' narrative to consider the fact you're full of it?

Strawman. My point was that dedicated science forums do not look like this, particularly professional ones, and there's a reason for that. SD.net does 'what are the implications of this new published research' about as well as Slashdot (which is to say, ok as idle chat, but not comparable to the professional debates that actually set policy). IMHO this is not a good place for highly speculative theory development and I don't think discussing the relative merits of bleeding edge unproven science paradigms is going to work very well.

[SirNitram]

Why do you pretend that flamewars, and SWvST are all that's here?

Strawman. I said that flamewars and making fun of morons are what this forum is particularly good at.

If that were really true, you'd have no objection whatsoever; just because we're better at one thing than another does not invalidate our ability to have enlightened conversations on dozens of topics.

Are you really so blind that you've seen none of the debates about real world issues and the like,

Of course not. I've posted in a few. This forum is comparable to most others of similar size and quality in that respect. That sort of discussion is fine, it's entertaining (hopefully) and sometimes enlightening.

or is it too disruptive to your 'You're all extremists who only exist to flame' narrative to consider the fact you're full of it?

Strawman. My point was that dedicated science forums do not look like this, particularly professional ones, and there's a reason for that. SD.net does 'what are the implications of this new published research' about as well as Slashdot (which is to say, ok as idle chat, but not comparable to the professional debates that actually set policy). IMHO this is not a good place for highly speculative theory development and I don't think discussing the relative merits of bleeding edge unproven science paradigms is going to work very well.

Highly speculative, unproven science isn't going to be useful outside of the lab or research paper, and frankly, that's for a reason. To pretend that there should be people as obviously grossly underinformed as colbert there spewing about it is certainly not helping the case.

[Rye]

So if utility dictates whether discussion of something will be welcomed here, what's wrong with telling someone they may have more luck elsewhere?

[Darth Wong]

Nothing really, except for the implication (quite possibly unintentional) that coberst actually had something worthwhile to say and we are just not capable of appreciating it. The fact is that coberst's ramblings are the worst sort of vacuous doublespeak, full of faux profundity but totally lacking in substance.

[Starglider]

If that were really true, you'd have no objection whatsoever; just because we're better at one thing than another does not invalidate our ability to have enlightened conversations on dozens of topics.

The general culture on this board is to attack incoming arguments with the debating equivalent of a machine gun. If there are any bullet-proof truths in there they gets a trial by fire and come out vindicated. Crap gets blown away. Occasionally valid arguments get taken out simply because their supporters don't have the debating skill to defend them and the odd salvageable poster who managed to look like a troll in their initial posts bites the dust. This collateral damage is acceptable given that this is neither a support group nor a forum for making critical real-world decisions, and this style avoids wasting time and makes for great entertainment.

Scientific debate is similar in the sense of ruthlessly chipping away bogus arguments over a long enough time scale (months to decades, depending on the field). On a comparable time scale though it's quite different. Generally things stay polite and careful to make sure things aren't dismissed prematurely, to allow for the fact that explaining complex theories is really hard and misunderstandings are common, to allow time for independent replication. In particular constructing theories is a brainstorming process that progressively adds pieces to an initially fragile structure, which doesn't have a chance of surviving the machine gun treatment until it's been developed, tested, tempered and generally made fit for public consumption (if it ever reaches that stage). For another thing debate on speculative theories necessarily involves a lot of probability, fuzzy ('possibly', 'probably', 'strongly suggests', 'weakly confirms' etc) and explicit ('observed correlation of x', 'evidence confirms to 90% likelihood' etc), which you don't see here for both style reasons and because the focus is on solid facts and extremely likely theories (likely enough that the probabilistic verbiage is redundant).

It can be irritating when scientists get impatient and push their underdeveloped ideas out in a popsci book, particularly if it includes self-important philosobabble (cognitive scientists and physicists are particularly prone to this). Regardless, this board focuses on the 'blast any perceived flaw' style (other than in fanfics) of debate and while that's great for many purposes (and in short supply on the post-usenet Internet), it just isn't appropriate for some things. If this particular poster actually had any worthwhile speculation, my advice to take it elsewhere would be sound.

Highly speculative, unproven science isn't going to be useful outside of the lab or research paper, and frankly, that's for a reason.

Research papers are what you get after the peer review process. Workshops, conferences, coffee room discussions, the entire sci.* newsgroup hierarchy and tens of thousands of mailing lists and web boards are what you get before it. And no, these aren't intended for general consumption nor is it useful to participate unless you have a specific interest in or use for untested and half-formed theories.

To pretend that there should be people as obviously grossly underinformed as colbert there spewing about it is certainly not helping the case.

True, sorry, guess I cut that one way too much slack. A persistent problem with open science forums is that you get random trolls in a way that never happens in faculty break rooms or conferences, yet people start by according them the same respect they'd give a colleague. On the AGI list all kinds of crackpots come on, but all they get is prodding to 'please be more specific' and 'cite some sources' and 'so how do you account for studies X, Y and Z' until it's utterly clear that they're worthless (and even then they tend to get ignored rather than booted unless they're persistently spamming). But this is SD.net and I should get out of the habit of being so indulgent, regardless of whether I'm already considered harsh elsewhere.

[Darth Wong]

Research papers are what you get after the peer review process. Workshops, conferences, coffee room discussions, the entire sci.* newsgroup hierarchy and tens of thousands of mailing lists and web boards are what you get before it. And no, these aren't intended for general consumption nor is it useful to participate unless you have a specific interest in or use for untested and half-formed theories.

Silly me, I thought that research papers came from research projects conducted in labs, rather than a bunch of people brainstorming on the Internet.

[Starglider]

Silly me, I thought that research papers came from research projects conducted in labs, rather than a bunch of people brainstorming on the Internet.

They do, but said people do actually like to talk to each other about what we're working on, thus the existence of forums to do so. Some important breakthroughs in AI have come from workshops (e.g. the seminal Dartmouth conference), mainly because it's so cross-disciplinary, but then plenty of important work has been done by lone geniuses as well. It's true that every so often a group of cranks join up and try the 'brainstorm an AI design over solely a discussion group' idea, generally with hilarious results. I don't know if this happens in other fields, but I get the impression that it's responsible for train wrecks like the electric universe <strike>paradigm</strike> cult.

[Ritterin Sophia]

Humans are meme (idea) adoring creators.

I've read these pages, but this sticks out to me.....

[aerius]

The fact is that coberst's ramblings are the worst sort of vacuous doublespeak, full of faux profundity but totally lacking in substance.

I was wondering why I was having flashback of the pseudo-intellectual crap from The Matrix movies.

[Rawtooth]

The fact is that coberst's ramblings are the worst sort of vacuous doublespeak, full of faux profundity but totally lacking in substance.

I was wondering why I was having flashback of the pseudo-intellectual crap from The Matrix movies.

It wasn't apparent to you from the start?

And I have to ask, especially since I'm considering double-majoring, is all psychology about as worthless as whatever coberst is saying?

[brianeyci]

Rawtooth, depends if your parents are rich. If your parents are well off and willing and able to bankroll your future well into your thirties, study whatever you want.

If you need money, forget psychology. In our university psychology is one of the largest classes, and supply and demand means once you come out, your degree's worthless.

At least pick a humanities that nobody is going for. Something with a direct link to some kind of career. Don't study the same stuff blond bimob's studying. At least blond bimbo's got nice tits and ass, you probably don't so you won't survive a psychology degree unless you go masters and doctorate.

[Mobiboros]

And I have to ask, especially since I'm considering double-majoring, is all psychology about as worthless as whatever coberst is saying?

No. It's not. I majored in psychology and while I didn't go into the field (You need a Masters if not a PhD to actually work in the field and I was just fed up with school), I know many people who continued on. Psychology has some very useful areas. The value of real therapy as well as research is there.

This comes witha HUGE however. However it's VERY hard to seperate out the good from the absolute crap. There's no middle ground. What's good is very good. The advances and uses of medicines to help people suffering from true chemical imbalances as well as stress induced trauma/anxiety has helped quite a lot of people. And the research into neuro-psych, social-dynamics, computational psych and industrial psych (Psychology of work environments. Very close to ergonomics but focused on human behavior) will become extremely useful as they mature.

The problem is what's bad. Decades of excruciatingly bad theories, compounded by an almost religious adherance to outdated ideas. Anyone with a half assed idea that makes people temporarily feel better and blame others for all their problems can easily get a book published or go on a talk show. Add in that the majority of the public distrusts anyone who tries to figure out the human mind because they feel we're absolutely too complex to be understood. Add in, yet again, that the field is flooded because most schools undergrad degrees in psych are really just a liberal arts degree with classes specialized in wanking about the incorrect theories of dead guys who did way too much drugs.

So basically the answer is. Psychology can and will be useful, but you need to really dig to make it so.