I am not looking for a straight answer to this question but some references on how I should further investigate this problem.


  1. Let us consider a person bob
  2. When Bob makes a statement, he should know each element (words) he is using in his sentence.
  3. If Bob is making a sentence where he doesn't know some of the words he is using. Then for Bob, it's not possible to make sense of the sentence he made.
  4. Now assume that Bob tries to make a sentence "I don't know X".
  5. If Bob's sentence is true, then he doesn't know what X is. This means Bob doesn't know what he is saying.
  6. Else, if his statement is false, then it's OK.

Where it leads me is when I say, "I don't know X", it's either false or I have no ideas what I am saying.

This is perplexing and leads me to how a person can know what he doesn't know.

  • Without a prior theory of " knowing", that is , a criteria of what it means to know something, knowing becomes simply a value judgement rather than fact judgement. Jun 3, 2023 at 10:27
  • 1
    You are playing with words... and that is not philosophy. Jun 3, 2023 at 12:39
  • 3
    You are mixing two uses of word "know": 1) to use a word, e.g. "water", in a sentence, and 2) to explain e.g. with chemistry theory, what water is. Jun 3, 2023 at 12:41
  • Let's say X_w:= just the written word and X_c: is the concept of X. Also, there is an association from X_wX_c. Through this, I get what you mean by saying there are two use of the word "know". Thanks, @MauroALLEGRANZA. But, can you give me some references where peoples deal with this type of thing? Jun 5, 2023 at 7:00
  • Muchas gracias for the question. Hints at a familiarity with philosophical techniques. That outta the way, I don't encourage a dive down this rabbit hole. Jun 5, 2023 at 7:06

3 Answers 3


In a more pragmatic sense, if someone says that they don't know X, what they usually mean is one of the following:

  • They have never heard of X before (they can't retrieve an entry for X in their brain's "database")
  • They have heard of X but never studied the concept (they have an entry for X in their brain's database, but the entry is pretty much empty)
  • They have heard of X, they even studied the concept to some degree, but in so doing realized that X has pointers to a bunch of related, more advanced concepts and topics, all of which would need to be studied and understood in order for them to say that they properly know X.

For example, I have an entry in my brain for "String Theory", and I've read some summaries, so I "know" that it points to a bunch pretty advanced concepts in Physics and Math, which I haven't studied yet, so I "know" that I "don't know" String Theory.

In short, if you define knowing X as being able to find entries for X (and related concepts included in X's definition) in a given database, then it is certainly possible to check whether X is "known" or "not known" in that database. If you see your brain as a biological database, this pragmatic definition should work, as it should with a database in the more conventional informatics sense, or the parameters of an artificial neural network storing the knowledge learned by an AI agent like ChatGPT, etc.


Before they discovered the Americas, Europeans did not know what lay beyond the Atlantic. But they did know how to find out. So, in one way they did not know what lay beyond the Atlantic. But what was known (part of the Atlantic) was the basis for discovering something "completely" unknown. It seems paradoxical, but it isn't really.

Take another example - Before the planet Uranus was discovered, nobody knew it was there so nobody was looking for it. It was discovered accidentally. It was completely unknown. Nobody could ask (let alone answer) any question about it.

However, various irregularities were observed in the orbit of the planet Uranus. The irregularities enabled mathematical calculations of its position. Finally, in 1846, Neptune, the eighth planet from the Sun was discovered. What astronomers initially discovered by accident enabled them to identify and discover something previously unknown. What we know can be the foundation of a new question and new knowledge.

It is true that you have to know something in order to ask a question. The answer fills in a gap in your information. You can't ask a question unless you know enough to ask it (and enough to recognize an appropriate answer).

The distinction between known unknowns and unknown unknowns was made famous in 2002 by Donald Rumsfeld when he was Secretary of Defence in the USA. But the distinction was first made in 1955 by two American psychologists, Joseph Luft and Harrington Ingham, in their development of techniques of analysing intelligence information.

I can't give you a strictly philosophical reference for this, but you may find this Wikipedia entry helpful Wikipedia entry - There are unknown unknowns.


The Flammarion engraving -The Flammarion Engraving, first documented in Camille Flammarion's 1888 book L'atmosphère : météorologie populaire

We live in a world full of errors & uncertainties, but that doesn't make knowledge impossible. The real problem is our intuition that 'true' always means getting to look behind the curtain - 'getting our homework marked by god'. When we think hard though, we find 'true' isn't in the 'facts', it is in the whole situation of evaluating them, and never stand separately from them.

-from this answer: Why is a measured true value “TRUE”?

I like the example 'opposites', of how we develop and use a 'language game', from simple to learn examples to very abstract cases: Life and Death as one and the same? The pairs are not unambiguous though, and the intuitiveness of the game can often cause us problems and contradictions. I like the elaboration of how we do this kind of thing, in Surfaces and Essences: Analogy as the Fuel and Fire of Thinking by Douglas Hofstadter and Emmanuel Sander.

Wittgenstein used the example of the word 'game' itself to demonstrate language-games, drawing attention with it to how a word can be intuitive and useful, even while we can't exhaustively and precisely define it. We feel we know people are engaged in a game when we see it. I feel he crystalises his point here:

"In this sort of predicament, always ask yourself: How did we learn the meaning of this word ("good", for instance)? From what sort of examples? In what language-games? Then it will be easier for you to see that the word must have a family of meanings."

-Wittgenstein, in Philosophical Investigations

The Hilbert Programme to provide a finitistic proof of the consistency of the axioms of arithmetic, Lord Kelvin's alleged declaration in 1900

“There is nothing new to be discovered in physics now. All that remains is more and more precise measurement.”

-supposedly from Royal Institution lecture Nineteenth-Century Clouds Over the Dynamical Theory of Heat and Light 27 April 1900, but at best a gross paraphrase

and the modern discovery that we don't know what 95% of the cosmos is made out of, illustrate how our greatest progress has been from revelations of the scale of our ignorance, and as we have thought our job of understanding was nearly done we broke out into new vistas of uncertainty.

"I don't know X"

Consider the game of pointing, and a child using a finger in that game, to identify something as unknown to them. True? False? Meaningless? Those labels are irrelevant, because their use of the pointing game is likely effective. They have conveyed something, likely to someone that can give them information. How will the witness to the pointing answer? Many contextual cues and clues will be involved, in deciding what 'knowing' will be in this case. Likely they will both get meshed into the 'Why?' game, discussed here: "Why ask why" and its scions

Vervaeke has this nice term 'cognitive grip', that we look for handles on our knowledge of the world, that can help us manipulate that knowledge effectively, towards acting on the world. He relates that to the idea that knowledge is about building up 'salience landscapes', where we gain useful new information, by arranging facts/data/observations relationally between themselves and to ourselves. For instance, by situating ourselves in a meaning-cosmology, like the 'Why?' game leads to. Discussed more in this answer: According to the major theories of concepts, where do meanings come from?

We can't know what we don't know. But, we can situate ourselves effectively towards that:

"I seem, then, in just this little thing to be wiser than this man at any rate, that what I do not know I do not think I know either."

-Socrates, in Plato's 'Apology'

Rumsfeld's famous declaration "There are unknown unknowns" is another example of such a situating. Science is always tentative, and in such a cautious and humble approach, we have found our deepest understandings.

  • Hilbert Program + Lord Kelvin??? Maybe some intermediate statement is missing... Jun 3, 2023 at 14:15
  • @MauroALLEGRANZA: I felt they are good examples of people who thought they knew what they didn't know
    – CriglCragl
    Jun 3, 2023 at 14:22

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .