Why MBTI is a fraud

[Tannhauser]

After a few posts in Architect's "MBTI takedown" thread, I began thinking about the paradox that MBTI seemingly makes surprisingly precise descriptions of individuals once you know his/her type, while at the same time not being a scientific theory.

First of all, in my opinion it is easy to show why MBTI is not a scientific theory: it does not make falsifiable statements. It says a lot about what people tend to do, but it never makes a statement which can be used to prove the theory wrong. For example saying that INTPs tend to be logical thinkers is not going to be invalidated by finding someone who scores as INTP but is a very 'unlogical' thinker. There are many obvious ways in which the theory can explain itself out of such a miss.

So I thought: Knowing it is not scientific, how come then, that it is so useful in predicting people's behaviour?

The answer hit me when I thought about the difference between, say, Einstein's Relativity, and the MBTI: Relativity made predictions which were almost absurd given what our experience told us. For example, it predicted the bending of light in gravitational fields, which was very counter-intuitive considering light does not have mass. This turned out to be a correct prediction and thus corroborated the theory. In other words, the theory didn't try to fit observed facts into it but instead deduced them, and even deduced facts which were not even imaginable before the theory. Now, MBTI differs in a fundamental way: it only recycles what we already observe and categorises that into concepts. It has never produced a prediction which was not already apparent.

In other words, MBTI is almost like a tautology: I can observe that some person is an introverted, logical person, and then I say "since you do Y, that means you are type X". Then we extrapolate what these introverted, logical people typically do, and then say "since you are type X, you tend to do things Y". And this is exactly what makes it so deceptively precise. Note that this is similar to what a psychic does: he/she makes you reveal things about yourself to him/her, and then feeds back the information to you in different form.

 

[Architect]

it does not make falsifiable statements. It says a lot about what people tend to do, but it never makes a statement which can be used to prove the theory wrong

You make too big a point of it. This is a behavioral theory, we don't hold it to the same level of falsibility as a physics theory! The Big Five theory is the same. It has a huge amount of research and data behind it, but a person that falls into any of the categories won't act that way robotically, all the time.

With psychological theories you have to relax the conditions a little bit, otherwise you wouldn't have any theories at all. As you note people act capriciously, so by your definition no theory would suffice. Really surprises me what you guys expect sometimes, but then people act illogically and capriciously, and out of their type on occasion.

Besides which that is easily accounted for by making it a probabilistic theory. It isn't just because nobody has been willing to spend the millions to do the data collection.

 

[Brontosaurie]

I agree with the basics of what you are saying. To me, type is not a theory but a convenient terminology for something about human nature. The innovation isn't scientific, but philosophical - meaning it doesn't create knowledge, but makes some aspect of intuition available in discourse. Still we all have very different interpretations. It's barely burgeoning yet.

Type is inadequate for making predictions, but very useful in getting to know a person.

 

[Architect]

Type is inadequate for making predictions ...

of an individual, but is very accurate for making predictions of a population. This is what it means to be a statistical theory.

Example, go find a million INTP's and put them on a basketball court. With MBTI you could accurately predict how most of them would react, and of course there would be outliers who would not fall into that prediction.

 

[Brontosaurie]

of an individual, but is very accurate for making predictions of a population. This is what it means to be a statistical theory.

Example, go find a million INTP's and put them on a basketball court. With MBTI you could accurately predict how most of them would react, and of course there would be outliers who would not fall into that prediction.

You may be correct, although it seems likely that a more scientific instrument such as the Big 5 has better predictive power in cases where one is interested in populations. I don't see this as the main domain for typology, since only description and correlation are requisite. Typology has its strength in aiding personal reflection by explicating core human values, not as a scientific instrument. Certainly i would love for typology to inspire or progress into science, and it's not impossible, but we cannot call it science just because there's a bogus test that makes fraudulent claims of being scientific. If you strip away the non-science psycho-metaphysics of MBTI, all that's left is an unreliable test instrument with far less empirical validation than Big 5, that posits a bimodal distribution which cannot be found in reality with the test as research tool.

 

[Architect]

You may be correct, although it seems likely that a more scientific instrument such as the Big 5 has better predictive power in cases where one is interested in populations.

Possibly, but if so it gets there by sacrificing complexity. The Big Five is so stripped down as to be almost completely useless, and it's hardly a predicative theory as MBTI is.

I agree with your sentiments about MBTI for personal reflection, but I think it also has a sleeper role as a scientific theory. Why it's not seems to be historical accident.

 

[Brontosaurie]

Possibly, but if so it gets there by sacrificing complexity. The Big Five is so stripped down as to be almost completely useless, and it's hardly a predicative theory as MBTI is.

I agree with your sentiments about MBTI for personal reflection, but I think it also has a sleeper role as a scientific theory. Why it's not seems to be historical accident.

Sleeper role? Sure. I can trace the potential convergences every day. It's a source of fascination. In its current state though, the MBTI is considered scientifically irrelevant for good reason.

 

[onesteptwostep]

Yeah the Big Five just says 'you have this much xxx and this much xxx', that's about it. It doesn't provide any descriptors beyond that. There's not much utility in the relational aspect. MBTI is connected to a lot of other personality/behavioral/cognitive theories too, which is why (well one of the reasons why) it's useful.

 

[Architect]

In its current state though, the MBTI is considered scientifically irrelevant for good reason.

I think it traces back through two main schools of thought in psych which is Freud versus Jung.

 

[Puffy]

I think it traces back through two main schools of thought in psych which is Freud versus Jung.

Both Freud and Jung are thought to be irrelevant in modern psychology though? Most psychology professors treat the field's origins as an embarassment you briefly mention then move swiftly on from. (= Big paranoia about coming across as 'scientific'.)

Sorry if I'm mistaken, but I'm not of the impression that MBTI is treated seriously in contemporary psychology and has a fringe position in relation to it.

That in itself doesn't dismiss it obviously. There's a lot about modern psychology I don't like, and a lot in Freud and Jung I find interesting. I just mean that if you're framing it in terms of Freud vs. Jung that's likely to be considered contentious.

 

[Architect]

Both Freud and Jung are thought to be irrelevant in modern psychology though

Yes agree. And I'm not an expert in this, but my belief is that modern psyche is Freud derivative, even though they've abandoned Freud in favor of pseudo attempts at scientific rigor. I think they threw out the baby with the bath water, i.e. 'holistic' psychology in favor of attempted scientific rigor.

Most practitioners are aware of MBTI (we know several shrinks) and some more or less use it, but they mostly put in the realm of hippie psychology.

 

[Analyzer]

I agree with the basics of what you are saying. To me, type is not a theory but a convenient terminology for something about human nature. n.

This reminds me of what Ludwig von Mises distinguishes as theory vs history.

“Thymology is a branch of history or, as Collingwood formulated it, it belongs in ‘the sphere of history.’ It deals with the mental activities of men that determine their actions. It deals with the mental processes that result in a definite kind of behavior, with the reactions of the mind to the conditions of the individual’s environment. It deals with something invisible and intangible that cannot be perceived by the methods of the natural sciences. But the natural sciences must admit that this factor must be considered as real also from their point of view, as it is a link in a chain of events that result in changes in the sphere the description of which they consider as the specific field of their studies.

Theories are deductive based models while thymology/history/psychology are inductive based fields which is capable of changing as new input is added.

 

[RaBind]

I think they threw out the baby with the bath water, i.e. 'holistic' psychology in favor of attempted scientific rigor.

I remember my psychology teacher talking optimistically about an anticipated paradigm shift, where the focus would move to more holistic perspectives in examining psychology. However I think psychology will always have a disposition towards behavioral/functional (in the same sense as how black box testing is understood in computing) perspectives because behaviors are the most basic easily observable empiric evidence, this is also a crutch in favor of the pesudo-scientific perspective.

I also think that focusing on treatments to alleviate symptoms discourages tackling the task of attempting to fully understand psychological phenomena. It encourages directing the focus of the field towards practical rather than theoretical work, analogous to an extent to the role of engineering compared to more theoretical research. I guess I'm saying that armchair theorizing should be encouraged more, as long as the theorist know what they are talking about, which I know is a somewhat disagreeable and underrated opinion.

Theories are deductive based models while thymology/history/psychology are inductive based fields which is capable of changing as new input is added.

Yea this is very much similar if not exactly what I'm talking about. An analogy might be how theoretical mathematics works compared to evidence based science, where theories are based heavily on and around empirical evidence.

 

[Tannhauser]

You make too big a point of it. This is a behavioral theory, we don't hold it to the same level of falsibility as a physics theory! The Big Five theory is the same. It has a huge amount of research and data behind it, but a person that falls into any of the categories won't act that way robotically, all the time.

With psychological theories you have to relax the conditions a little bit, otherwise you wouldn't have any theories at all. As you note people act capriciously, so by your definition no theory would suffice. Really surprises me what you guys expect sometimes, but then people act illogically and capriciously, and out of their type on occasion.

Besides which that is easily accounted for by making it a probabilistic theory. It isn't just because nobody has been willing to spend the millions to do the data collection.

Certain fields of psychology, for example the work of Tversky and Kahneman on cognitive bias, are highly scientific in the sense that they make concrete, testable hypotheses. They never tried to create a theory for why these biases exist, however. They only described the way they actually influence behaviour.

MBTI seems to be of a completely different sort altogether: it seemingly creates a framework which seeks to explain why people act as they do, meanwhile never making any testable hypotheses. And from what I understand, the reason it appears to work is almost like a trick of the mind: It groups the population into 16 types based on behavioural traits, then tests which of the 16 fits best for an individual, and then basically tells the individual 'what he is' based on the way that type was defined in the first place.

But I could do the same trick with an arbitrary grouping of behavioural traits: Let's say we make it simple and have 2 groups: group A are those who tend to be more angry than average. Group B are the opposite. Then I take an individual and test which group he fits in, let's say it was A. Then, as if by magic, I tell the individual that he tends to be more angry than average. Same trick as MBTI does, except it invokes an intricate 'explanation' based on 'functions' etc.

Maybe that is why it has such a magical effect when you read your own profile for the first time: it tells you what you already know but possibly didn't put in clear terms before.

 

[Black Rose]

@Tannhauser

What you describe as how the MBTI is tautological must have first come from empirical observations. Because anger is a real biological mechanism but before neuroscience we did not know its location being in the amygdala. So to begin with you must have observations to present as evidence that a phenomena is present. The mechanism is important but only after observation. From a website I discovered Jung had an IQ of 160. Smart people usually make better observations and create theories that way. Only when it comes to people whom are capable of distinguishing the differences observed will there be the capacity of forming a well developed life experience of them. So being that the MBTI practitioners have the life experience and the training to recognize traits such as introversion and extroversion, they do not rely completely on the paper tests results. They had to observe hundreds of examples of types that online tests do not qualify as giving thorough results. It has to be a person to person life experience of reciprocal communication.

Raising children also makes it possible to understand the development of the psyche from the internal perspective. If you know why a child does what they do you become aware of the empathic connection which is more than behavioral. Some people do not have that empathic connection so they cannot see things from the inside perspective of the other person. To understand the psychological type you need to have that inside perspective to know what is happening in their head and heart and that does not come from a paper test. The reason MBTI is not a scientific theory is because Theory of mind is not a scientific theory. Nor is meta cognition a scientific theory. Both are necessary to understand type. And by having someone experienced with both to guide you in learning how they apply social reinforcement among practitioners deepens those understandings between older and younger students. That is why amatures fail at typology so often, they do not have social reinforcement from experienced individuals.

 

[Tannhauser]

@Tannhauser

What you describe as how the MBTI is tautological must have first come from empirical observations. Because anger is a real biological mechanism but before neuroscience we did not know its location being in the amygdala. So to begin with you must have observations to present as evidence that a phenomena is present. The mechanism is important but only after observation. From a website I discovered Jung had an IQ of 160. Smart people usually make better observations and create theories that way. Only when it comes to people whom are capable of distinguishing the differences observed will there be the capacity of forming a well developed life experience of them. So being that the MBTI practitioners have the life experience and the training to recognize traits such as introversion and extroversion, they do not rely completely on the paper tests results. They had to observe hundreds of examples of types that online tests do not qualify as giving thorough results. It has to be a person to person life experience of reciprocal communication.

I am not sure if I understood your point, but I disagree that you would have to know the neurological mechanism of anger before being able to categorize a person as angry. That experiment could have been performed even before people knew what a brain was. And this is also why MBTI is much less impressive than it looks: it merely takes what we observe and tells it back to you in different words.

 

[Architect]

it seemingly creates a framework which seeks to explain why people act as they do, meanwhile never making any testable hypotheses.

I'm not getting through to you. I'll make it clear one last time

• MBTI is about hidden variables (Functions or Preferences) which are observed through high level behaviors which are influenced by many factors
• Like any hidden variable theory then demonstration is achieved via statistical methods

Put simply, an INTP acts like an INTP, most of the time. And what it means by 'acts like an INTP' will also vary by individual. However there is a commonality between INTP's as they seek to express their 'hidden variables'.

This is no different then a million other theories, like quantum mechanics. I can't tell you what one particle will do, but I can tell you what aggregates will do. It's really that simple, if that doesn't get across then there's not anything more I can say, this is all standard science.

If you doubt the data veracity get the MBTI manual, it's filled with tables and studies.

 

[Black Rose]

I am not sure if I understood your point, but I disagree that you would have to know the neurological mechanism of anger before being able to categorize a person as angry. That experiment could have been performed even before people knew what a brain was. And this is also why MBTI is much less impressive than it looks: it merely takes what we observe and tells it back to you in different words.

Any theory begins with taxonomy. If no one was ever angry the trait of anger would have been impossible to link to its mechanism. And if MBTI was a fraud would not this mean that its taxonomy would be random? Because there is a reliable taxonomy of behaviors then this means there must be a mechanism for it just as we discovered the mechanism for anger. How is that a fraud? Are some people making unsupported claims about what MBTI can do?

Fraud = Deliberate Deception

 

[Tannhauser]

Any theory begins with taxonomy. If no one was ever angry the trait of anger would have been impossible to link to its mechanism. And if MBTI was a fraud would not this mean that its taxonomy would be random? Because there is a reliable taxonomy of behaviors then this means there must be a mechanism for it just as we discovered the mechanism for anger. How is that a fraud? Are some people making unsupported claims about what MBTI can do?

Fraud = Deliberate Deception

That is my point -- MBTI is nothing but a taxonomy. However, it purports to explain behaviour by an elaborate theoretical framework. Meanwhile, if we strip away all the statements which only amount to mere speculation, for example the speculations about inferiors, extroversion/introversion etc, we are left with nothing but a grouping of people into groups of similar behaviours based on arbitrary criteria. So the taxonomy is not random, but it is arbitrary.

The utility of MBTI is clear: it can help you to remind you of your own tendencies, insofar as they pertain to the arbitrary taxonomy of the MBTI. However it has absolutely nothing to do with gaining insight into human psychology. In that regard, MBTI is as useful as the grouping of people into blondes, brunettes and red heads.

So we could have designed a new, arbitrary MBTI-like system like this:
- Create 16 different groups characterized by certain traits of your own choosing.
- Observe what people with these traits tend to do
- Create a questionnaire to reveal the tendencies of an individual so that you can assign a type to that individual
- Now you can come up with any 'theory' you like of why the grouping is at it is. You can invoke all sorts of ambiguous and speculative concepts like introversion/extroversion, functions -- say, shyness-functions or love-functions or thinking-functions etc.

 

[Tannhauser]

I'm not getting through to you. I'll make it clear one last time

• MBTI is about hidden variables (Functions or Preferences) which are observed through high level behaviors which are influenced by many factors
• Like any hidden variable theory then demonstration is achieved via statistical methods

Put simply, an INTP acts like an INTP, most of the time. And what it means by 'acts like an INTP' will also vary by individual. However there is a commonality between INTP's as they seek to express their 'hidden variables'.

This is no different then a million other theories, like quantum mechanics. I can't tell you what one particle will do, but I can tell you what aggregates will do. It's really that simple, if that doesn't get across then there's not anything more I can say, this is all standard science.

If you doubt the data veracity get the MBTI manual, it's filled with tables and studies.

Yes, you have raised the probabilistic point before. However, I think we differ on what a probabilistic theory is. It is not a theory which is "sometimes correct", it is a theory which generates precise probabilistic statements, for example "event E will happen with probability 0.15". From what I can see, MBTI is exactly a theory which is only sometimes right.

It is statistically verifiable only in the sense that someone who is typed as an INTP will most probably remain an INTP in the near future. That is no different than saying that someone who is shy will probably behave in a shy manner in the near future.

 

[Architect]

However, I think we differ on what a probabilistic theory is. It is not a theory which is "sometimes correct", it is a theory which generates precise probabilistic statements, for example "event E will happen with probability 0.15". From what I can see, MBTI is exactly a theory which is only sometimes right.

Mixing metaphors, conflating 'correctness' and probability. What is incorrect here? Take a gander at Personality Types in Software Engineering, isn't it making precise predictions? Just a small sample, some 100 people, but this predicts that you will find an INTP in software engineering at 2.5x the rate at which you'll find them in the general population.

I'll guess you're having trouble with the descriptions, such as "INTP's like logic", without an accompanying specific prediction "90% of the time INTP's will counter a discussion with a logical rebuttal". Well, so what? The data doesn't exist because nobody has paid for it. But one billionaire could easily spent a small fortune collecting such data, and over millions of INTP's would find that > 50% of the time, INTP's counter arguments with logical rebuttals.

 

[Hadoblado]

@Archie

I think it traces back through two main schools of thought in psych which is Freud versus Jung.

Yes agree. And I'm not an expert in this, but my belief is that modern psyche is Freud derivative, even though they've abandoned Freud in favor of pseudo attempts at scientific rigor. I think they threw out the baby with the bath water, i.e. 'holistic' psychology in favor of attempted scientific rigor.

Most practitioners are aware of MBTI (we know several shrinks) and some more or less use it, but they mostly put in the realm of hippie psychology.

Jungian psychology shows more in common with Freud than modern psychology. Both theorists fall into psychoanalytics which is only one of six major schools in psych. It seems like you’re saying that because Freud and Jung are two different schools of thought that can be thought of as being opposed to each other, that any opposition to one must come from the other. So the reason that Jung is rejected by modern psychology is because of Freud’s influence, when Freuds work is equally rejected by modern psychology for the same stated reasons: it’s unscientific.

• MBTI is about hidden variables (Functions or Preferences) which are observed through high level behaviors which are influenced by many factors
• Like any hidden variable theory then demonstration is achieved via statistical methods

But afaik current statistical evidence has only asked questions that lead to answers with no power to support the notion of functions.

The way I see it, if you want to contend that MBTI meets a relevant statistical standard while doing more than overlapping with what Tannhauser correctly describes as a taxonomy, the hidden variables you speak of should form the basis of a prediction that surface characteristics cannot merit alone.

So basically, can you make a statistical prediction from function that cannot be predicted by answers to questions of the ilk you find on the actual MBTI test?

If it is not possible to infer anything from function that is not also inferred from surface characteristic, then this is strong evidence that the functions themselves are immeasurable and therefor we have no reason to continue to assume their existence.

Is there any evidence for function?

You're acting like Tannhauser has not understood that psychological inquiry of this nature requires statistics and all the assumptions that come along with it. It is not the case that just because a standard of proof resides in statistics that this standard is automatically met by whatever pet theory takes haven on its uncharted shores. Evidence that people in a logic intensive field can be described as logical does not support a distinction of Ti/Fe vs. Te/Fi, or any other functional pair.

 

[Tannhauser]

So basically, can you make a statistical prediction from function that cannot be predicted by answers to questions of the ilk you find on the actual MBTI test?

Very nicely put. I've got nothing to add.

 

[onesteptwostep]

I don't see how that paradox discredits its utility.

The problem here is just epistemological. I find it somewhat disheartening to see empiricism taking over the discussion. The qualia of the functions are not quantifiable, sure, but most of the human condition aren't either. The strength of the MBTI is its universality in describing the different modes of the cognitive beings. It's not supposed to be empirical in any sense, it's supposed to be descriptive. The 'evidence' for the functions comes from the consensus of experiences from those who understand the theory. What makes happiness happiness? What makes sadness sadness?

 

[Tannhauser]

I don't see how that paradox discredits its utility.

The problem here is just epistemological. I find it somewhat disheartening to see empiricism taking over the discussion. The qualia of the functions are not quantifiable, sure, but most of the human condition aren't either. The strength of the MBTI is its universality in describing the different modes of the cognitive beings. It's not supposed to be empirical in any sense, it's supposed to be descriptive. The 'evidence' for the functions comes from the consensus of experiences from those who understand the theory. What makes happiness happiness? What makes sadness sadness?

Well, the paradox was: how does it seemingly generate precise statements despite being based on a completely speculative framework. As we have seen in this thread, the answer is that it actually doesn't generate any statements at all. Whatever it outputs is already what you have put in.

For example Architect proposed a statistical utility: if you know someone is INTP, we can predict that he will tend to think logically. But this is circular reasoning: in order to be typed as INTP you need to tend to think logically in the first place.

So what utility does it actually have? I am not sure at this point. Maybe help you to categorise your own tendencies, but then it is hard to see why you should use the arbitrary framework of the MBTI to do that. Would it not be better to look at your own tendencies as particular to you, without trying to force them into an arbitrary framework of functions, extroversion/introversion etc?

 

[Miss spelt]

Its utility has always only ever been to improve interpersonal relationships and understand your fellow man, his motivations etc.

Acceptance.

 

[onesteptwostep]

Well, the paradox was: how does it seemingly generate precise statements despite being based on a completely speculative framework. As we have seen in this thread, the answer is that it actually doesn't generate any statements at all. Whatever it outputs is already what you have put in.

It doesn't make statements which are bound empirically, sure.

For example Architect proposed a statistical utility: if you know someone is INTP, we can predict that he will tend to think logically. But this is circular reasoning: in order to be typed as INTP you need to tend to think logically in the first place.

Eh, I don't follow you here- I'm pretty sure that isn't circular reasoning.

So what utility does it actually have? I am not sure at this point. Maybe help you to categorise your own tendencies, but then it is hard to see why you should use the arbitrary framework of the MBTI to do that. Would it not be better to look at your own tendencies as particular to you, without trying to force them into an arbitrary framework of functions, extroversion/introversion etc?

Yeah, this is how Jung probably started on his 'types'.

 

[Tannhauser]

Eh, I don't follow you here- I'm pretty sure that isn't circular reasoning.

Whatever you would like to call it, it is saying "people who tend to think logically tend to think logically".

 

[onesteptwostep]

How does that account for Ne-aux, Si-tert, and Fe-inf?

 

[Tannhauser]

How does that account for Ne-aux, Si-tert, and Fe-inf?

It amounts the same thing: As long as you can categorise a tendency, you can test for the tendency, and then 'predict' they have that tendency.

 

[Hadoblado]

Its utility has always only ever been to improve interpersonal relationships and understand your fellow man, his motivations etc.

Acceptance.

I find it teaches you the tools required to properly engage with the subject of personality. Even if the conclusions are wrong, thinking in terms of differences is immensely useful for understanding other people. Just off the bat, introversion vs. extroversion explains so much conflict. In fact, regardless of empirical basis, I find all the dichotomies have helped me develop socially. Before exposure to MBTI, people who were different to me made no sense, and I felt forced into either thinking I was always wrong (depressing), or that everyone else was (arrogance).

How does that account for Ne-aux, Si-tert, and Fe-inf?

It doesn't because they aren't evidenced (thus far). By not accounting for them doubt is cast on their very existence.

Robert's logical, creative and quiet -> Therefore Robert is an INTP -> Therefore we can predict that he will be logical, creative, and quiet.

In the above example the category INTP is conceptual creep. It gives nothing while complicating the issue. Functions are a part of MBTI that should not fit this criteria. If the functions predict nothing that isn't evident from the information used to type a person, then that would prove that MBTI is more than a taxonomy. I think this paragraph from the OP gave a good example:

Spoiler

The answer hit me when I thought about the difference between, say, Einstein's Relativity, and the MBTI: Relativity made predictions which were almost absurd given what our experience told us. For example, it predicted the bending of light in gravitational fields, which was very counter-intuitive considering light does not have mass. This turned out to be a correct prediction and thus corroborated the theory. In other words, the theory didn't try to fit observed facts into it but instead deduced them, and even deduced facts which were not even imaginable before the theory. Now, MBTI differs in a fundamental way: it only recycles what we already observe and categorises that into concepts. It has never produced a prediction which was not already apparent.

So relativity predicted things that no other theory or intuition could and then those predictions were found to be true. In MBTI, the dichotomies used to categorise are clearly taxonomical and quite possibly arbitrary. You can categorise people using any set of terms that are reducible to approximate dichotomies. The reason MBTI is supposed to be different is that these dichotomies are the tip of the iceberg for the underlying functions.

If the dichotomies do not predict the functions then the theory is false, but if the dichotomies predict everything the functions predict then the functions are redundant, and the theory can be stripped back down to a self-evident taxonomy. So in order for the theory to hold merit, it must be that the dichotomies predict the functions, and then the functions must predict behaviour that the dichotomies alone do not.

The problem here is just epistemological. I find it somewhat disheartening to see empiricism taking over the discussion. The qualia of the functions are not quantifiable, sure, but most of the human condition aren't either. The strength of the MBTI is its universality in describing the different modes of the cognitive beings. It's not supposed to be empirical in any sense, it's supposed to be descriptive. The 'evidence' for the functions comes from the consensus of experiences from those who understand the theory. What makes happiness happiness? What makes sadness sadness?

The problem is not 'just' epistemological. There is a clear attainable standard here that will disarm this attack on MBTI. How about we circumvent epistemological debate by accepting we disagree on the importance of science, but devote this topic to whether or not MBTI falls within the realm of what should be accepted science. That way anyone who is disheartened by all this talk of empiricism can be reheartened. It's worth noting that these unquantifiable qualia such as happiness and sadness are far more measurable than type, even if it is indirectly.

I think typology-apologists (say seven times real fast) should focus on the distinction of perceiving versus judgement, which is not self-evident but derived from function itself. That we are able to type INTJ from INTP is only made possible by this additional distinction: a difference derived from function alone (thanks Cheese). I know that if someone came at me with that counter-example I'd be forced to move my goalposts

 

[onesteptwostep]

It's worth noting that these unquantifiable qualia such as happiness and sadness are far more measurable than type, even if it is indirectly.

I'll go further with this. In the camp of the qualia or emotion known as happiness, what differentiates gladness with joy? Delight with excitement?

Also, the MBTI is not prescriptive, it's descriptive. I wouldn't say 'prediction' either, more like, 'possibility'. The cognitive processes are (or either fall in one of the functions) the same, but the objective output of those process's, could be different.

 

[Call Me Fishmael]

It amounts the same thing: As long as you can categorise a tendency, you can test for the tendency, and then 'predict' they have that tendency.

I think you just summarized the scientific method, just in more vague terms. However, if the point is that the method does not work without a high degree of specificity, then that point is well taken. The question here is why we see this tendency, Ti-Ne-Si-Fe, as opposed to something like Ti-Ne-Ni-Te or Ti-Ne-Se-Fi? Is it just a case of confirmation bias, that we see this pattern because it was suggested? How do we determine the role of bias in a self-reporting system?

The interplay of the functions is what separates Socionics or MBTI from Big 5 and depression inventories and other descriptive tools and theories. If you get into Socionics, the theory is even more complex and specific than MBTI, making it more testable. Not to a scientific degree, but to a more precise one.

Whatever you would like to call it, it is saying "people who tend to think logically tend to think logically".

Except that it's not. Looking at it from the perspective of function axes, it's more like saying "people who tend to think deductively tend to struggle with producing an emotional atmosphere and interpreting the feelings of others," which is not evident unless you suppose the the issue as two sides of a dichotomy. Likewise for the other functions. It is predicting reactions to stimuli which are not present based on those which are present, thus delineating a correlation (or lack thereof) between the two which would not be readily apparent. In a way, you test whether something works as a proxy, a reverse-proxy, or neither. We ask "does Ti work as a reverse-proxy for Fe?" In my experience this has been true, as well as for the other function axes. It has helped me understand other people, and I have made what I deem to be accurate predictions of people based on the characteristics of the system. Isn't that the point?

 

[Hadoblado]

I'll go further with this. In the camp of the qualia or emotion known as happiness, what differentiates gladness with joy? Delight with excitement?

Seems like a derail. Make a (or find one of many already existent) thread(s) about it and many people will be more than happy to discuss epistemology with you.

The topic is interesting and warrants discussion, it's just you tend to change the subject instead of progressing a conversation (no offense).

Also, the MBTI is not prescriptive, it's descriptive. I wouldn't say 'prediction' either, more like, 'possibility'. The cognitive processes are (or either fall in one of the functions) the same, but the objective output of those process's, could be different.

How can it possibly be only descriptive when it describes a population? Intrinsic in categorising something is the prediction that like is as alike. If I describe you and your neighbor as INTP's and there is nothing you have in common, my description is not useful.

 

[onesteptwostep]

Oh no, I'm trying to allude to an analogy, it's not a change of subject.

It doesn't describe a population, it describes cognitive functions and their orderings. You would have to give a test out to find out whether one would be INTP in the first place to prescribe it to er, a group of people. That isn't prescription.

 

[Jungle]

I think the beauty of MBTI is that it is not objectively verifiable but it is subjectively verifiable in quite a striking way. What I mean is that an individual who has a high capability for personally derived logical reasoning (i.e. Ti) can test the system for themselves to see if it works. MBTI is a system that can be rigorously tested by Ti (subjective logic) but not by Te (objective logic).

If you try explaining this to a Te dom they will hit you over the head with the concept of confirmation bias. However, a Ti dom has the rare ability to cut through confirmation bias because they are constantly scrutinising new information as it comes to hand and checking whether it is consistent with their previous conclusions. On the other hand a Te dom will be very reluctant to accept information which contradicts a previously held understanding because it undermines the integrity of their primary function - therefore confirmation bias is a very real problem for them when they are working 'off the grid', i.e. outside the realm of externally verifiable facts.

I have no doubt in the veracity of MBTI because I have tested it against my observations of pretty much everyone I have met since I started studying it, as well as mentally reviewing previous relationships, and I have never found a personality who is not accurately and comprehensively described by one of the 16 types - including atypical individuals (such as an introvert in a leadership role who pretends to be an extrovert) and observations of the more complex dynamics like inferior and trickster functions. I am personally confident that my deductions are correct and that I have not been influenced by confirmation bias. However there is no way I can objectively prove that to be true.

If you are relying on the MBTI quiz to identify type you will have a hard time testing the system itself, even on a subjective basis. Because how are you are ever going to meet enough people who casually mention their MBTI test result to you? And what if their answers to the questions were distorted by their inferior function? The only way to thoroughly test the system is to get good enough at typing that you can confidently type almost everyone you meet without talking to them about type. You might need to pick a nominal type for them to begin and then observe them over time to see whether their behaviour is consistent with your hypothesis. If need be you can reclassify them and then check all the historical data (Si) against your new hypothesis. The way you know the system is valid is if, after mastering the skill of typing, you never ever have to retype someone after you have spent more than a few hours with them.

From a Te perspective that must sound like an insane whirlpool of confirmation bias. But from a Ti perspective it ends up being an incredibly innovative, accurate and useful system. I think it is worth noting here that the majority of influential MBTI theorists have themselves been INTP. It is a Ti-centric system.

In terms of usefulness it helps you to understand the people you share your life with. Nothing more and nothing less.

 

[Architect]

Jungian psychology shows more in common with Freud than modern psychology. Both theorists fall into psychoanalytics which is only one of six major schools in psych.

I'm not an expert in the history of Psych so thanks for the clarification.

It seems like you’re saying that because Freud and Jung are two different schools of thought that can be thought of as being opposed to each other, that any opposition to one must come from the other. So the reason that Jung is rejected by modern psychology is because of Freud’s influence, when Freuds work is equally rejected by modern psychology for the same stated reasons: it’s unscientific.

My impression was that they were more or less opposed

But afaik current statistical evidence has only asked questions that lead to answers with no power to support the notion of functions ...

Completely agree. My point is that people are throwing out the baby with the bath water. MBTI clearly needs more work, but instead of figuring out how to fix the issues most frequently people abandon it in favor of a new system (a-la Pod'Lair). New theories in science don't throw away old ones that work, they improve them (i.e. GR reduces to Newton's theory of gravity).

Saying MBTI is a fraud takes that path. I'm saying that instead, I believe if we recast it as a hidden variables theory it would keep the best of the old and recast it in a rigorous and verifiable manner.

This is a similar solution as Einstein did with GR. He recast gravity as a geometrical problem.

 

[Tannhauser]

I think you just summarized the scientific method, just in more vague terms. However, if the point is that the method does not work without a high degree of specificity, then that point is well taken.

I completely disagree that such a process has anything to do with science, the reason being exactly what I described in the original post: scientific theories and (useful) models generate statements which are not already presupposed as observed facts. If the theory/model cannot do this, it has no reason to be proposed in the first place. If it actually does generate such statements but the statements are wrong (which is a much better scenario), it can be falsified and we can proceed further to come up with better theories/models. And that is exactly why science, as opposed to mere speculation, or fitting observations into frameworks, is such an effective and powerful process.

Except that it's not. Looking at it from the perspective of function axes, it's more like saying "people who tend to think deductively tend to struggle with producing an emotional atmosphere and interpreting the feelings of others," which is not evident unless you suppose the the issue as two sides of a dichotomy. Likewise for the other functions. It is predicting reactions to stimuli which are not present based on those which are present, thus delineating a correlation (or lack thereof) between the two which would not be readily apparent. In a way, you test whether something works as a proxy, a reverse-proxy, or neither. We ask "does Ti work as a reverse-proxy for Fe?" In my experience this has been true, as well as for the other function axes. It has helped me understand other people, and I have made what I deem to be accurate predictions of people based on the characteristics of the system. Isn't that the point?

I think the points you are making about how MBTI works only describes the ideal it tries to live up to, not the reality of it. It is true that some of the statements of MBTI are derived from its theory about functions. However, if the statements cannot be interpreted in any concrete way, the theory is as useful as astrology. For example: MBTI tells us that the combination Se/Ne is impossible. But can we find a person which appears to have this combination from the way these concepts are described? Of course, because they have no clear definition.

 

[Inquisitor]

Spoiler

I think the beauty of MBTI is that it is not objectively verifiable but it is subjectively verifiable in quite a striking way. What I mean is that an individual who has a high capability for personally derived logical reasoning (i.e. Ti) can test the system for themselves to see if it works. MBTI is a system that can be rigorously tested by Ti (subjective logic) but not by Te (objective logic).

If you try explaining this to a Te dom they will hit you over the head with the concept of confirmation bias. However, a Ti dom has the rare ability to cut through confirmation bias because they are constantly scrutinising new information as it comes to hand and checking whether it is consistent with their previous conclusions. On the other hand a Te dom will be very reluctant to accept information which contradicts a previously held understanding because it undermines the integrity of their primary function - therefore confirmation bias is a very real problem for them when they are working 'off the grid', i.e. outside the realm of externally verifiable facts.

I have no doubt in the veracity of MBTI because I have tested it against my observations of pretty much everyone I have met since I started studying it, as well as mentally reviewing previous relationships, and I have never found a personality who is not accurately and comprehensively described by one of the 16 types - including atypical individuals (such as an introvert in a leadership role who pretends to be an extrovert) and observations of the more complex dynamics like inferior and trickster functions. I am personally confident that my deductions are correct and that I have not been influenced by confirmation bias. However there is no way I can objectively prove that to be true.

If you are relying on the MBTI quiz to identify type you will have a hard time testing the system itself, even on a subjective basis. Because how are you are ever going to meet enough people who casually mention their MBTI test result to you? And what if their answers to the questions were distorted by their inferior function? The only way to thoroughly test the system is to get good enough at typing that you can confidently type almost everyone you meet without talking to them about type. You might need to pick a nominal type for them to begin and then observe them over time to see whether their behaviour is consistent with your hypothesis. If need be you can reclassify them and then check all the historical data (Si) against your new hypothesis. The way you know the system is valid is if, after mastering the skill of typing, you never ever have to retype someone after you have spent more than a few hours with them.

From a Te perspective that must sound like an insane whirlpool of confirmation bias. But from a Ti perspective it ends up being an incredibly innovative, accurate and useful system. I think it is worth noting here that the majority of influential MBTI theorists have themselves been INTP. It is a Ti-centric system.

In terms of usefulness it helps you to understand the people you share your life with. Nothing more and nothing less.

MBTI is objectively verifiable and subjectively verifiable. There are many studies which provide data in support of the MBTI dichotomies as reckful put it. Therefore it can be useful in choosing a career for instance, if we simply look at where many INTPs tend to congregate. For example: MBTI INTP Career Report. The idea is that if you look at enough people with identical dichotomies, you can then draw conclusions about what kind of fields/careers are most likely a good fit. I have to say, though, that after reading reckful's posts, I'm inclined to stop conflating MBTI with type dynamics. I have not abandoned the idea of the functional stack, but after reading Jung, I'm not entirely convinced it's possible to elucidate a functional stack for each of the 16 MBTI types. For instance, in Psychological Types, Jung states a Ti dom's inferior functions are all extraverted in attitude, which means the stack should be reframed as Ti-Ne-Se-Fe. Reckful thinks otherwise...

I completely disagree that such a process has anything to do with science, the reason being exactly what I described in the original post: scientific theories and (useful) models generate statements which are not already presupposed as observed facts. If the theory/model cannot do this, it has no reason to be proposed in the first place. If it actually does generate such statements but the statements are wrong (which is a much better scenario), it can be falsified and we can proceed further to come up with better theories/models. And that is exactly why science, as opposed to mere speculation, or fitting observations into frameworks, is such an effective and powerful process.

I think the points you are making about how MBTI works only describes the ideal it tries to live up to, not the reality of it. It is true that some of the statements of MBTI are derived from its theory about functions. However, if the statements cannot be interpreted in any concrete way, the theory is as useful as astrology. For example: MBTI tells us that the combination Se/Ne is impossible. But can we find a person which appears to have this combination from the way these concepts are described? Of course, because they have no clear definition.

At the end of the day, and this is what Jung more or less argued, you either give primacy to the object or to the subject. If the fact that the functional stack theory/framework lacks empirical data to support it rests uneasy with you (ie you relate primarily to the world through what is objective), then you're obviously not going to embrace the functions. OTOH, if the lack of data doesn't present a problem in your psyche (ie you give primacy to the subjective in your life), then you're more likely to enjoy viewing the world through the functional stack lens.

At the end of the day, the only thing that will satisfy you is hard data. Unfortunately, it seems that at least for the functional stack, that data is missing/incomplete/unsupportive. However, as reckful said, the dichotomies do have some empirical evidence. So if you're uncomfortable with the idea of the functions (which you seem to be), then I might suggest just viewing people in terms of which preferences seem to dominate in that individual (S or N, I or E, etc.).

 

[Call Me Fishmael]

Jungle put it so elegantly, and quite well. Keep Ti and Te in their proper spheres.

I would also argue that Ne and Se are clearly defined, at least in contradistinction to each other. I refer to functions because, at this point, I'm much more invested in Socionics, seeing as it's a more complete system. Non-functionalized MBTI is not worth the time.

 

[cheese]

I think typology-apologists (say seven times real fast) should focus on the distinction of perceiving versus judgement, which is not self-evident but derived from function itself. That we are able to type INTJ from INTP is only made possible by this additional distinction: a difference derived from function alone (thanks Cheese). I know that if someone came at me with that counter-example I'd be forced to move my goalposts

Just want to clarify in case this is unknown to readers:

J/P is determined by which function you extravert the most.

So an INTP, despite being a dominant 'judger' (Ti is a judging process), is a P because Ne (a perceiving process) is his highest extraverted function.

Whereas an INTJ, despite being a dominant 'perceiver' (Ni dom) is a J because Te (a judging process) is his highest extraverted function. NiTeFiSe

[For extraverts, the dominant function *is* the most extraverted function.
Eg ENTJ: TeNiSeFi]

Essentially, the J/P dichotomy is derived from (a small component of) function theory.

 

[Idunno]

quick question to tannhauser, if MBTI is fraud, yet has been used since 1920's, where 3 of the most famous psychologists thought of the same thing, "blindly" and is still being used today, what is true in personality theory? Nothing?

If you can answer this simple question, you got me convinced

 

[Miss spelt]

Just want to clarify in case this is unknown to readers:

J/P is determined by which function you extravert the most.

So an INTP, despite being a dominant 'judger' (Ti is a judging process), is a P because Ne (a perceiving process) is his highest extraverted function.

Whereas an INTJ, despite being a dominant 'perceiver' (Ni dom) is a J because Te (a judging process) is his highest extraverted function. NiTeFiSe

[For extraverts, the dominant function *is* the most extraverted function.
Eg ENTJ: TeNiSeFi]

Essentially, the J/P dichotomy is derived from (a small component of) function theory.

I still can't accept this as universally valid.

Perhaps in some types. In general, however, I think MBTI has it wrong.

oh well. I already made a thread about it.

 

[Inquisitor]

I still can't accept this as universally valid.

Perhaps in some types. In general, however, I think MBTI has it wrong.

oh well. I already made a thread about it.

It occurred to me that another way to think about this whole issue is to stop believing that there is only one reality: the objective reality. Instead, the world for all of us comprises the external/objective reality as well as an internal/subjective reality. The MBTI measures only the external, but the internal world remains largely hidden away from its probes. In fact, I would venture to say that all psychological tests, at best, can only glean a very indirect picture of an individual's subjective reality and are overwhelmingly concerned with comparing externally visible differences between people.

Reckful, being an INTJ, and true to what type dynamics would predict, approached this whole issue inductively. If you look at his posts, he has beautifully stated his case by supporting it with an exhaustive collection of data, references to the literature, and articles. For him, the logical sequence starts from the external and moves towards an internal, subjective possibility of what type dynamics should be. Reading his posts, I was struck by the lengths he went to base absolutely every assertion on objective evidence. There was not one single reference to his own personal experience. He strove to be 100% true to the data, and not leave anything unsupported hanging out there. You can see that the judging for him is directed entirely at the objective evidence, and not at all at his own subjective impressions.

Here's the leap that he makes: He basically says that an MBTI P-dom individual must necessarily have a dominant perceiving function. If the objective/external reality were indeed the only reality for all of us, he would be correct, but....it is only one of two realities. And because of this, MBTI fails to take into account the subjective/inner world of the individual, because quite frankly, it's too difficult to measure/quantify.

At the end of the day, what type dynamics predicts, (and this is why it does pass the test for me as a predictive theory of human behavior), is that if you are externally/objectively observed to be a P-dom, you must necessarily be subjectively J. Since testing is unable to directly reveal the individual's subjective reality, type dynamics fills that gap.

Bottom line: if you believe that the only reality is that illustrated by the data, then type dynamics is flawed; but if you believe as I do (and as Jung did as well) that there exist two realities, then type dynamics helps illuminate the inner world that is unquantifiable.

 

[Hadoblado]

quick question to tannhauser, if MBTI is fraud, yet has been used since 1920's, where 3 of the most famous psychologists thought of the same thing, "blindly" and is still being used today, what is true in personality theory? Nothing?

If you can answer this simple question, you got me convinced

Appeal to tradition
Appeal to celebrity
Appeal to normality

Sorry, a bit blunt I know. What I'm trying to say is that none of these argument forms are valid. If you want an example of a belief that meets all these criteria that isn't true, choose a religion you don't believe in and tell me why it's wrong.

 

[Miss spelt]

Bottom line: if you believe that the only reality is that illustrated by the data, then type dynamics is flawed; but if you believe as I do (and as Jung did as well) that there exist two realities, then type dynamics helps illuminate the inner world that is unquantifiable.

How can anyone disagree with this? You can't. It's a good, clean thought. Respect.

However,
I have to point something out.

I think you have taken your turn at misinterpreting extraversion and introversion as it applies to thinking,

using reckful again as an example, I have this to say.

The primary difference in attitude as it pertains to the thinking function is whether the reasoning process starts with the subjective, briefly appeals to the objective idea, and returns once more to the subjective... or if its continual purpose is always to appeal to objective ideas at the expense of subjectivity.

I think that this particular poster has exemplified introversion in his thinking.

I understand that he refrains from discussing personal experience and the like,
but he is also impenetrable in his resolve. He appears to be well rehearsed and to have a complete understanding of the theories as he discusses them, no doubt, however he shows no allegiance to any one thing that doesn't make perfect sense, to him.

This individual is 100% organized in his thoughts and 100% confident in his assertions, and I think he's a perfect example actually of one who starts his reasoning with the subjective, appeals to "the data" or generally held notions, freely dismisses that which he deems inconsistent and freely synthesizes those which make sense to him, and then returns to his own subjective interpretations as supreme.

I think, with regards to introversion as seeking depth, we have a great example of someone who penetrates very deeply into thought, examines every idea for its merit, and callously discards whatever does not fit within his framework.

It can be argued that it's Ni at play to some extent, but that is no less consistent with the approach I'm taking anyway.

 

[Idunno]

Appeal to tradition
Appeal to celebrity
Appeal to normality

Sorry, a bit blunt I know. What I'm trying to say is that none of these argument forms are valid. If you want an example of a belief that meets all these criteria that isn't true, choose a religion you don't believe in and tell me why it's wrong.

xD. Its wrong because the amount of made up altercations of a single story, seen throughout history. The logical fallacies, the biological implications. The oblivious appeal to the human psyche.

Bary Greenstein- If you beat everyone and we find out there is a cheat in this thing, you shouldn't be surprised that people are thinking that you cheated, even though you didnt. i think you have an idea of where im trying to go with this

 

[Hadoblado]

Nice

Reasons that would perpetuate the momentum of MBTI include it's accessibility, the money-train it facilitates, the identity people have tied up in the belief (more-so than other belief systems), the competence that people want to keep (feels bad to have one of your major skills rendered redundant), people hard committing to the belief (difficult to admit you're this fundamentally wrong), the actual good it does for people's understanding despite being wrong itself, and the feel-good it enables.

It's conceptual crack, and we're all addicts.

 

[reckful]

Here's the leap that [reckful] makes: He basically says that an MBTI P-dom individual must necessarily have a dominant perceiving function.

I've only made eight posts here so far, Inquisitor, and in three of them — two of them replies to you — I've noted that I agree with James Reynierse that the so-called cognitive functions are a "category mistake." I don't subscribe to any functions model, I don't believe in "dominant" functions, and I don't believe in characterizing S and N as "perceiving functions" or T and F as "judging functions."

 

[Idunno]

Nice

Reasons that would perpetuate the momentum of MBTI include it's accessibility, the money-train it facilitates, the identity people have tied up in the belief (more-so than other belief systems), the competence that people want to keep (feels bad to have one of your major skills rendered redundant), people hard committing to the belief (difficult to admit you're this fundamentally wrong), the actual good it does for people's understanding despite being wrong itself, and the feel-good it enables.

It's conceptual crack, and we're all addicts.

Haha that's funny. So do you think it's all fake? If you were to give me a percentage of its validity after stating your reasons of its preservation what would you say?