Now, you seem to be implying that even though recognition is not measurement, each recognition is made on the basis of having made a measurement. However, even that is not obviously true - what do I measure when I recognise a spelling mistake? I see the mistake, but seeing isn't always measuring.

what do I measure when I recognise a spelling mistake — MetaphysicsNow

In order to determine wether there is a 'z' or an 's' written, you measure several things. Once that is done, you can compare it with what shape ought to be written. Children learning to write are not as skilled in this as you probably are, and thus more often confuse the two, resulting in more spelling mistakes. Even to determine wether we describe something as a curve or an angle, we need to measure. Hence it's so much harder to make an ai that can accurately recognize handwriting, than to make an ai that can accurately write handwriting.

In order to determine wether there is a 'z' or an 's' written, you measure several things.

That's disputable. For instance, I simply saw that you misspelt "whether" "wether" - I didn't measure anything and I did not even compare your "wether" with a correctly typed "whether". Developing the skill of spotting spelling mistakes may or may not involve "measuring" , although it would be a strange definition of "measuring" if it did, but even so that would not in the least entail that everytime I spot a spelling mistake now that I have that skill, that I am measuring something. When you learn to play a musical instrument such as a violin, you begin by concentrating very hard on where you place your fingers on the fingerboard. When you are a proficient violinist you no longer need to do that. Things that are done in order to gain a skill are not necessary to continuing to manifest that skill.

All I know about IQ is that all the RAF could ever find to say about me was that I was extremely intelligent (it's doubtless gone off since I took the test!). They key point about all this inequality drivel, in my view, is that before we start using such things to make some people important, we must decide what human beings are for, as we might decide, for instance, that a knife is for cutting, at which point sharpness matters. Since humanity seems to be for nothing in particular, it is an end in itself, as are we all, and judging people on size of ears. range of voices or ability to pee far should be worth what they deserve - nothing.

Okay, but I don't understand why you assume that intelligence CANNOT be measured. It certainly can.

If you are right, then the word "intelligent" would never have exited, much less used by anyone. It is because we have some (vague) concept of intelligence that we can use the word. I don't need to know the precise definition to say that Richard Feynman was intelligent. I don't need to know the precise definition to say that Albert Einstein was intelligent. But objectively speaking, both of these people are very intelligent academically. The fact that we are saying this already proves that we have some concept in our mind that is capable of testing people's intelligence. So intelligence can be tested qualitatively. The only question is how quantitatively. IQ tests are merely one of these approaches to quantitatively measure intelligence.

Now I understand that you do not understand statistics. If you don't understand statistics, then you won't even know what IQ tests are about. Why are you arguing if you don't know IQ tests?

We do have a standard for both volume measurement and iq measurement to compare it too. — Tomseltje

You keep saying this, but when you are asked what that standard is, you demur or insist that the measurement is the standard.

Simple, we have different kind of iq tests. Had all been 100% accurate, there would be no difference. However, when we use different tests, the results differ, hence either one of the tests used is inaccurate, or both are. — Tomseltje

No, that won't work either. If intelligence is just what the tests measure, and you insist that this is the case for all tests of intelligence, then different results can only mean that intelligence is different in each case.

Tomseltje, you should understand by now that you cannot cheat your way out with this simple maneuver of equating intelligence with test results. Even setting aside the issue of accuracy, suppose we accept your idiosyncratic definition of intelligence - what then? So you have a device that measures something, and all we know about that something is that it is just what the device measures.



If you want to have a substantive discussion, you have to address the question of what intelligence is, and how intelligence tests can measure it, how accurate and how useful such tests are, etc. But for that you actually have to care and know something about the subject, and I don't think that you do.

That there is a concept of intelligence is not in question and how we gain that concept, or any other for that matter, is perhaps an interesting question. However, that the concept corresponds to a measurable property of human beings is what is in question. Human beings deal with plenty of complex concepts that we learn how to use without having to measure anything (at least in any straightforward sense of "measure"): intelligence is one of them, niceness is another example, there are plenty of others. Somewhat akin to 17th century physicists who developed the (now discarded) theory of phlogiston, psychologists and others seem to presume that there must be something objective and measurable underlying the use of the concept (yes, they even do this for niceness apparently). But that is simply a presumption and is entirely unsupported. Furthermore, unlike the developers of phlogiston theory, the motivation for the development of IQ testing and the surrounding technical apparatus was not disinterested pursual of knowledge about the nature of our use of a concept, but was actually driven by prejudices about what kind of people had more of this magical stuff than others.
The burden of proof here is not on the skeptic who accepts we have an interesting concept but who suspects a category mistake is being made when the concept is assimilated to physical concepts such as heat and mass. The burden of proof is on those who insist that the concept corresponds to a measurable property. As @SophistiCat points out, the burden of proof is actually more specific than just identifying such an objective property, but also that the property comes in differing amounts and is not simply present or absent.

And the only thing that IQ tests have ever been able to tell about anyone is how good or bad they are at taking IQ tests. — MetaphysicsNow

This the wrong point. Why IQ tests are different to other psychological ones? The questioning of IQ validity is an evidence of low IQ.

You know, I'd never thought about it like that. By God you are right! I am a dumbass. Thanks for pointing that out.

Nicely put, and since you seem to have gained high marks in previous IQ tests, perhaps @Belter will take you more seriously than he or she is willing to take me.

By God you are right! — MetaphysicsNow

You can be sure. I work in a very related domain. People often do not know what is an IQ test.

The thing is the types of question in an IQ test are clearly not arbitrary. They are all of an immediately recognisable 'sort', vaguely - patter-recognition, sequence-recognition and sense-in-context. So if it's true that we can recognise intelligence in a fuzzy, family resemblance kind of way, then it must also follow that we could recognise the sorts of problems an intelligent person should be able to pass in the same vague, fuzzy way. Put a collection of these sorts of problems together and you have an IQ test, so it would measure intelligence according to our fuzzy definition by virtue of being made up only of questions which match that fuzzy definition.

If, on the other hand, we wish to deny any authority to our vague notion of the sort of question an intelligent person ought to be able to answer, then we must also discard the notion that we can recognise an 'intelligent' person by the sorts of things they are able to do.

Basically, we don't look inside anyone's head for intelligence. If we judge it at all we judge it by the things people successfully do. Put a series of those sorts of things in a test and, by default, you do indeed have a device for measuring the thing we're calling 'intelligence'. Either that, or admit that we really don't know what sort of thing an intelligent person should be able to successfully do, and so abandon the idea that we have any means of measuring it, neither intuitive nor quantitative.

I think the problem with intelligence testing is an entirely resolvable one. It simply doesn't currently test the same sorts of things that we normally associate with intelligence in common usage. It tests a very narrow range and places way too high a value on speed (which we hardly value at all in real life). All of these problems are resolvable. In fact a team at Cambridge Brain Sciences Unit are doing exactly that.

I respectfully disagree.

If you have a concept and can distinguish between an intelligent person or not, whether subjectively or objectively, you are qualitatively measuring intelligence. I can compare average Joe and Einstein and claim that Einstein is smarter, and Joe doesn't have to be severely unintelligent for me to say that. There is some concept within me and most likely most of the other people, that is capable of measuring intelligence. As such, intelligence is measurable, at least qualitatively.

IQ test is, to a certain degree, good measure of intelligence. Most people of Mensa International is indeed, both subjectively and likely objectively, smart. Also, IQ test is a statistical (relative) measurement, and not an absolute measurement.


Since you mentioned measuring physical concepts, I have to make sure you understand that any physically measurable phenomenon must first be defined prior to measurements. As a matter of fact, these definition could change (although it usually doesn't affect quantitatively because of the generality of physical axioms tend to be consistent or at least a good approximation of other physical axioms) depending on the axioms of the physical model it is based on. You mentioned mass and heat, but these first started as an intuition that things feel "heavy" when lifting heavy objects and that one feels "heat" when touching hot objects. The concept was always there but the problem was how to quantitatively measure them. Thus came the definition of mass and heat, which was based on intuition but is practical enough that people accept it.

For example, how would you define temperature? Definition of temperature is actually quite complicated than most people think, despite people generally accepting the concept and its measure. I have (physical) chemistry background, and thus have thermodynamic background when it comes to definition of temperature, but more fundamentally, it can come from statistical mechanics, which can also easily applied to quantum mechanics if the temperature is defined based on grand canonical ensemble. So the definition of temperature also changes depending on the particular physical model that it is based on.

Since the term "intelligent" is not universally well defined, their measurement is obviously not easy. IQ tests are one of these attempts to define "intelligent" so that measurement can be done.

Since the term "intelligent" is not universally well defined, their measurement is obviously not easy. IQ tests are one of these attempts to define "intelligent" so that measurement can be done. — FLUX23

In my opinion, this rules for all psychological (and scientific in general) concepts. Only logical and math concepts are "universally well defined".

That taking an IQ test is a manifestation of human intelligence is not something I have denied or would deny. What I am denying is that it follows from the fact that intelligence can be manifested in a variety of ways, including taking IQ tests, that intelligence is a thing which comes in amounts.

If, on the other hand, we wish to deny any authority to our vague notion of the sort of question an intelligent person ought to be able to answer, then we must also discard the notion that we can recognise an 'intelligent' person by the sorts of things they are able to do.

I'm not sure what you are implying here, but perhaps there's an argument to draw out of it. Suppose I accept that taking an IQ test is one manifestation of intelligence. Suppose I know someone who manifests other kinds of intelligent behaviour - she speaks a language, plays a musical instrument and has an active social life for instance, maybe she's also quite manipulative of others. Suppose that person fails miserably everytime she takes an IQ test of whichever variety. I might be surprised that she is bad at taking IQ tests, certainly, but then again perhaps not. The point is that there would be no (at least obvious) logical contradiction in supposing that she was simply bad at those tests yet still capable of manifesting all kinds of other intelligent behaviour.

Social, linguistic and musical features of intelligence are also measured in the IQ test. People only show the limitations of empirical science for psychology of intelligence. It is a clear case of selection bias.

Psychological, yes. Scientific in general, to a degree, yes. I agree.

I respectfully disagree.

About what specifically - the burden of proof charge?

If you have a concept and can distinguish between an intelligent person or not, whether subjectively or objectively, you are qualitatively measuring intelligence.

I've been down this particular path in this thread already - distinguishing one thing from another does not always involve measuring. At least, it is by far from obvious that it does. You may have an entire theory of cognition that is based on a representational theory of perception which assimilates all perceptual activity to measuring the environment, and so everything that involves perception in any way involves measurement, but it would be a whole different thread to examine that kind of idea and what might be wrong with it (the notion of "representation" is an often abused and confused one in the philosophy of mind and psychology).

OK, so I think the first question to answer here would be whether you think we generally consider someone who can solve complex maths puzzles and play a musical instrument as more 'intelligent' than someone who can only do one or the other. My feeling is that that is exactly the term we would use for such a person. So, following from this, so long as the sorts of problem included in the test cover a wide enough range of the sorts of tasks we consider an 'intelligent' person should be able to do, then the test could be considered to be measuring intelligence.

There is, of course, the question of whether such a range of tasks could ever be captured in a reasonably sized test, but I think if we were claim that they could not, then we'd be starting to open the definition of intelligence so wide as to make the word useless, certainly we'd end up defining it in a way it is rarely used. I even think its pushing the boundaries of normal use to describe a good artist as 'intelligent', though I agree it's borderline acceptable.

The second question is how we can relate our use of the expression "more intelligent" to your idea that maybe intelligence doesn't come in amounts.

We could say that the expression was simply a nonsensical one, but it seems a bit selective to accept our intuitive idea that there is such a thing as intelligence but then deny what appears to be an equally intuitive sense that some people have more of it than others.

Hence my first question. Does having an ability in more areas of intelligence make one more intelligent overall? If so, it seems to me to be eminently possible to measure the sum intelligence by some test or other. The problem is merely a pragmatic one of getting the test to match the things we use the word for.

I apologize that I haven't read the entire thread thoroughly.

I would greatly appreciate it if you could provide me the post where you explained why you think intelligence is not measurable. I probably missed a few posts or part of the post where you mentioned them, but all I could find is your assumption that they are not measurable, and not the justification of why it is so.

OK, so I think the first question to answer here would be whether you think we generally consider someone who can solve complex maths puzzles and play a musical instrument as more 'intelligent' than someone who can only do one or the other.

We're getting down into some complicated issues now! Well, personally I do not play a musical instrument. My colleague who sits next to me does. Other than that difference we have much in common educationally, and the similar jobs we perform we perform to all intents and purposes to the same degree of proficiency, advice seeking between us is a two way street. Do I consider myself less intelligent than he is on the basis that he plays a musical instrument, or did I just not have the same opportunities that he had growing up? Of course, I'm biased, so it's probably not for me to answer, but then why should the outcome of that question be based on our capacity to complete an IQ test either?

Of course, I've met stupid people, or rather people I would call stupid: voters of populist nationalist parties for instance, and so with people like that if I were asked if I were more intelligent, well, I might be inclined to say yes (although I've met some people who do vote that way who are remarkably good at mathematics). Here I think the example introduces the idea of there being some kind of value-based judgement going on in deciding whether someone is more or less intelligent than another.

So, to your general question

Does having an ability in more areas of intelligence make one more intelligent overall?

I suppose I have to answer "It's complicated".

No problem,take a look through the posts between myself and Tomseltje on the idea that recognising intelligence involves measuring intelligence. I start off with the example that I don't measure anything when I recognise someone's face in a crowd, so recognition does not always involve measurement. He/she disagrees and the exchange peters out with two ideas left hanging and unsettled: 1) That even if we might gain an ability through using measurement in the beginning, it does not follow that continual proficiency requires continually measuring, and 2) What I suspect is some kind of general theory of cognition lying in the background of Tomselje's approach that assimilates all perceiving to measuring.

You recognize someone's face because your brain can process your vision in which particular set of properties in the vision matches the "properties of a face" that you have and not that of the background. As such, you are incapable or will have difficultly in detecting camouflaged faces because it has the properties very similar to the environment and as such your brain processes the face as the environment.

This is a pure act of measurement. You are measuring face.

OK, so your claim that all recognition is measurement is based on a very specific theory of mind in which the term "measurement" has a very specific technical use, since in a non-technical sense I certainly do not need to measure anything when I recognise a face in a crowd. I can certainly imagine circumstances where I might do so - for instance, I have a picture of a crowd of people on the one hand, and a picture of a person to find in that crowd on another, and there I can see myself in some circumstances perhaps measuring the length of the persons nose in comparison to his brow and so on and comparing it with similar measurements of faces in the crowd. Also, if I have to recognise a specific person from a crowd of very similar looking people, I might go down to measuring (well, counting) how many freckles are on faces and so on However, that kind of measuring is certainly not what I am doing when I recognise my mother in a crowded supermaket. So if there is measurement going on - as you claim - it is in a very special and technical sense and in which case we have got about as far as we can go simply with the discussion of the validity of IQ tests, since it all seems to hang on a theory of mind. For sure, I think we are going to have significant differences of opinion on the mind-brain issue in general, given the differences of opinion about what IQ tests are measuring, but this is not the thread for that.

Well, the brain is actually "smarter" than that.

First, you don't need to recognize every single pieces of the face to recognize that it is a face. Your concept within your brain provides criteria for face recognition (which have been continuously refined ever since you were born and you opened your eyes).

Second, experience allows you to recognize certain faces faster than the other. With experience (for example your face, which you probably see very often in the mirror), your criteria is optimized for certain cases. For example, if you had tons of paint on your face, sometimes it is hard or takes time to recognize that it is you in the picture out of all the other people. However, you will eventually recognize yourself because you have other criteria to judge that your face is your face.

Somewhat rough but a lot of measurements are actually going on in your brain whether you realize it or not. This is particularly fascinating to me.

I suspect that the bias you mention is the main reason why people resist intelligence testing. They wish to retain their authority to determine people to be 'intelligent' or 'stupid' without having to have such use circumscribed by any kind of testing which risks settling the matter in a manner which may offend their purpose.

Personally, I agree that most populist nationalist voters are stupid, but then I'm convinced that they would fail any intelligence test one could devise, so this is not a cause for concern for me.

What I think bothers people most is the other end. The poet and the painter, the philosopher perhaps, don't wish to be tested in any way which may compare them to the mathematician and the scientist. I personally have no real concern here either because I really can't see a need for such a test.

Where I do take some offence is the current state where those with shall we say non-conventional intelligence, resist having their skill measured quantitatively and yet wish to maintain fervently their authority to do so in their own subjective manner even in public discourse. That seems, to me, a little disingenuous.

Where I do take some offence is the current state where those with shall we say non-conventional intelligence, resist having their skill measured quantitatively and yet wish to maintain fervently their authority to do so in their own subjective manner even in public discourse. That seems, to me, a little disingenuous.

OK, we may not have a disagreement about these kinds of people - it sounds like at some level they are being hypocritical, but if you had an example or two of who you are talking about it might be clearer to me what exactly the issue is here.

I'm not sure quite what examples might suffice. I suppose to support my argument I'd need quotes from the less mathematical/scientific community saying they don't want intelligence testing but they do know a couple of geniuses in their fields. I'm not sure I'm going to be lucky enough to find a quote that captures all that. It's more based on the regular use of terms like "genius" to describe people like Mozart and Shakespeare that would certainly not be used to describe Taylor Swift or JK Rowling despite the latter two having achieved just as much in their fields. There's a sense that there's definitely some objective thing that Mozart and Shakespeare have unarguably got, yet a reluctance when it comes to defining what it is.

At my university we used to inevitably have the odd dropout from the BSc courses, occasionally they would transfer to a BA. This was a relatively normal occurence. We would take bets, however, on any successfully going the other way.

Well, "What's a genius" probably deserves a thread of its own - but mere membership of Mensa doesn't cut it for me. We can probably say without courting too much controversy that Mozart is a better composer than Taylor Swift, and whilst we might get heckled by adolescents for suggesting so, we could probably find a musicologist who could give us all kinds of good reasons for believing that to be the case objectively. Similarly, perhaps some professor of literature can be appealed to to show us exactly how Shakespeare is a better writer than JK Rowling. What I'm still not seeing, though, is how admitting all that would support the idea that there is one measurable thing/property called intelligence that Shakespeare and Mozart had more of than do Taylor Swift and JK Rowling.