Is it a fallacy if someone claims they need an explanation for every word of your argument to the point where they don't understand common terms?

For example, suppose someone said, "If a dog bites people, then it's a vicious dog." Then someone counters, "What do you mean by 'bites'? Define 'biting people'. Define 'dog'."


10 Answers 10


A fallacy in an argument requires that there be an argument at all.

What you're describing:

Then someone counters, "What do you mean by 'bites'? Define 'biting people'. Define 'dog'."

is not even an argument. There is no fallacy there, because it doesn't even make a claim.

What it does – in the example you present – is bog down the discussion with unreasonable demands. This person is not honestly engaging in argument, they are trying to prevent the argument from progressing at all.

Asking for definitions is not in general dishonest. There are, of course, many terms more open to contradictory, valid definition, than “dog” and “bites”.

If you are asked to define terms like, for example, “person”, or “real”, or “harm”, then it's probably best to have some working definition of those terms explicit, in case the argument could hinge on the properties of such a term.


Generally speaking, in philosophical discussions, it is often required to provide definitions for words that seem obvious otherwise, let me give you an example:

When you say : The King of France is bold . Russell may ask you, what do you mean by The? (I will not talk about the theory of descriptions here).

And Russell's question here would be legitimate, and it is significant to analytical philosophy, in what sense do you claim that the king of France such and such...?! (and this word in Russell's terms means more than it seems to mean)

So, I would assume that an analytical philosopher is NOT committing a fallacy here, since I know that even the slightest of our common uses of language can still present some ambiguities at a deeper analytical level.

So, back to your example:

What do you mean by bite?

I do not know of a particular example of a discussion where 'bite' would pose any definitional issues, but consider this dialogue:

  • A: Your dog bites
  • B: What do you mean by bites?
  • A: I mean bite, you know what I mean.

In this situation, B must provide examples as to what definitional level he is talking about. And may reply:

B: I mean, in what sense my dog bites? in the aggressive sense of biting people to hurt them? or in the sense of playful biting?

As you can see here, there is still an ambiguity in the statement. But there are caveats, these factors can resolve the ambiguities, and add in implied meanings, even if A does not provide further details.

  • The tone and state in which A talks (unless they are texting)
  • B already knowing his dog is aggressive.

So ambiguities only arise either when B knows his dog is not aggressive (so he suspects that 'bites' means playful biting), and when they are texting (so B cannot see how A feels about it).

On the other hand, B's question "What do you mean by bites?" , is also ambiguous, probably more ambiguous that the word 'bite' itself.

Because, when one says "what do you mean by ...", or "how do you define..." , one asks a question that is as ambiguous as the word he asks the definition for.

And A should ask him what level he is talking about, in what sense he is talking, and what kind of definition he wants? and should ask for some examples, to which B will reply : "Do you mean a, b or c...?"

In case all ambiguities are resolved, but B is still asking questions about defining the word 'bite' , then you may refer to these fallacies:

  • Sealioning Asking questions and demands for endless answers.
  • It might be a Red Herring , if B drags the discussion towards 'bite' instead of talking about his dog's behavior.

As well as fallacies suggested in the comments above.

As Frank Hubeny's answer pointed out, sealioning is considered a pseudo-fallacy by Bo Bennett, Check this link

But remember, it might not be a fallacy if there are still ambiguities that need to be resolved, so that the parties know exactly what they are talking about.


A sentence is anything that is said, for example : Your dog bites people.

But in philosophical discussions, philosophers (especially analytical philosophers and logicians) deal with propositions.

You dog bites and Your dog bites are one sentence, that may refer to 2 (or more) propositions. Propositions are what you mean by a sentence, not what is articulated or expressed, but also what is implied.

Many sentences can refer to one proposition, like I love Paris, J'aime Paris and 我喜欢巴黎...etc.

And many propositions may be implied by one sentence, I love you is one sentence, that can imply many meanings (propositions) : Romantic love, parental love...etc.

  • 2
    Rather than adding infinite edit notes with changes at the end, you should edit the answer as a whole to stand as if it were always the best version of itself. Anyone interested in past versions can view the revision history.
    – V2Blast
    Commented Mar 3, 2019 at 0:14

As @SmootQ notes some names for such argumentation may be "sealioning" and "red herring".

Regarding sealioning, Bo Bennett describes this not as a logical fallacy, but rather as a pseudo-logical fallacy:

A subtle form of trolling involving “bad-faith” questions. You disingenuously frame your conversation as a sincere request to be enlightened, placing the burden of educating you entirely on the other party. This is not a fallacy; it is more of a form of deception. As always, be careful in assuming you know the other person’s intent. On the surface, “sealioning” looks a lot like legitimate and honest Socratic inquiry.

This would not meet all three of Bennett's characterizations of a logical fallacy although some might consider it a logical fallacy:

  1. It must be an error in reasoning, not a factual error.
  2. It must be commonly applied to an argument either in the form of the argument or the interpretation of the argument.
  3. It must be deceptive in that it often fools the average adult.

However, Bennett considers a red herring to be a legitimate logical fallacy:

Attempting to redirect the argument to another issue to which the person doing the redirecting can better respond. While it is similar to the avoiding the issue fallacy, the red herring is a deliberate diversion of attention with the intention of trying to abandon the original argument.

Bennett also cautions his readers against raising fallacy issues against their opponents' arguments:

I caution you against correcting fallacies that your opponent might raise. As you will see in this book, fallacies go by many different names, and there are varying definitions for the fallacies. Except for a handful of fallacies that have been around since the time of Aristotle, most fallacies are under a continual redefining process that might change the name of the fallacy or the meaning of the fallacy. The bottom line is to focus on exactly what error in reasoning you are being accused of, and defend your reasoning—not a definition or name.

Knowing about fallacies is mainly a form of self-defense, not a form of offense in an argumentation. If you point out a fallacy to your opponents, "what you certainly should be prepared for, is your opponents pointing out your fallacies, and if you know about fallacies, you will be ready to defend yourself."

Bo Bennett "Being a Smart-Ass" Logically Fallacious https://www.logicallyfallacious.com/tools/lp/Bo/LogicalFallacies/207/Being-a-Smart-Ass

Bo Bennett "Pseudo-Logical Fallacies" Logically Fallacious https://www.logicallyfallacious.com/tools/lp/Bo/LogicalFallacies/6/Pseudo-Logical-Fallacies

Bo Bennett "Red Herring" Logically Fallacious https://www.logicallyfallacious.com/tools/lp/Bo/LogicalFallacies/150/Red-Herring

  • @sdenham Asking for clarification is indeed a valid move, but the parties must get back to the original point to avoid Red Herring. On the other hand I did not yet check the example of red herring on bo bennett's reference to confirm as I am texting now from a smartphone.
    – SmootQ
    Commented Feb 27, 2019 at 18:41

Who and where did a person made such an argument? It seems like you encountered an internet troll or something.

You do such an approach:

  • If you honestly need to clarify EXACTLY what someone else said.
  • If you want to avoid to answer to an obviously good argument that devastated yours.
  • If you want to make fool of someone.
  • If you want to make a joke - troll somebody.

If someone does that with negative intentions, I first make sure they are just fooling around with me, then I reply " Stop answering my questions with questions. If you don't have an answer it's ok." If the person insists that they need to get an explanation of every word, then they are obviously not qualify for people that have answers to your questions. They need to learn the language semantics first.

Meanwhile, remember the first case. If you have a conversation about a philosophical question, then it IS valid to ask for a few clarifications.

"What do you mean by 'bites'?" Seems valid to me. You can bite to play, if you are a playful dog, or you can bite to harm, if you are a scared dog.

Define 'biting people'. Seems also a valid question to me. What people? Why bite them and how bite them, soft or to harm.

Define 'dog'." Seems stupid to me. In case I had to deal with that person, it would end there. Take it with a grain of salt. I don't know the context of the original conversation and you post on Philosophy.

I would answer: "Google is your friend." and disengage.

If someone wants to show off that way, just stop interacting. The person does not qualify to give you useful answers. Disengage and continue the conversation with the rest of the people that can be actual helpful. your time is too precious to be wasted at such games.

  • Absolutely. There are some who will even ask "What precisely do you mean by 'dog'?" without any actual goal of understanding.
    – user21820
    Commented Mar 3, 2019 at 3:16

It can heavily depend on the type of the argument and the words used. For example, in pseudo-scientific claims many scientific-sounding words are used which have nothing (or almost nothing) to do with the topic.

For example, someone wants to market a bogus "healing armband" to me, and claims that it uses "quantum-magnetic resonance to optimize the flow of energy through the body".

In this case it should be a valid argument to ask for the definition of every word in this sentence, if for not else, then in order to show that they have no idea what they are talking about.


On its own, I would not consider asking for such common place definitions to be a fallacy, but it could be a low quality argumentative technique if they're using it the wrong way.

On one hand, demanding definitions for every word is a known way to stall an argument until one party gets bored. Precisely defining all words is known to be an impossible task, so a rude debater can merely continue to demand you expound on the semantics of your statements (including word meaning) until the discussion unravels. As George Eco points out in his answer, if they start debating the meaning of "dog," then they probably need to explain why they think a precise definition matters... otherwise they're likely just trying to abuse the wording of the arguments.

On the other hand, sometimes words that you think have universally agreed upon meanings actually don't. Often there's merely substantial overlap between meanings. In these cases, a debater may be demanding definitions to point out the lack of agreement on meaning. I find this is terribly obvious in the ongoing social debates regarding abortion, which can all really be summarized with a single question "When does life begin?" Life is a funny word in that we have remarkable agreement on its meaning in general, but in particular corners of the space of questioning, the differences in opinions turn into remarkably large obstacles (in particular, the beginning and end of life are troublesome, as is whether criminals have a right to life after doing something egregious).

Potentially the most extreme version of this is the debate surrounding the word "God," with or without the capital "G." Not only do atheists not agree with theists, and theists from different religions disagree, but we often find meaningful and poignant disagreement between members of the same congregation!

If I may draw from a fiction author for a closing quote, here's a snippit from Stranger in a Strange Land by Robert Heinlein:

Short human words were never like a short Martian word — such as "grok" which forever meant exactly the same thing. Short human words were like trying to lift water with a knife.
And [God] had been a very short word.


I don't think this can be called a logical fallacy, because a fallacy is committed by someone who is presenting an argument. Demanding an explanation of every single term is not arguing, it's an attempt to avoid arguing.

Perhaps it would better to call it "gaslighting", seeing as how they may be trying to slip in the notion that you (the victim of this form of behavior) don't really know any of the things you profess to know.

Other possible terms are "being a contentious jackass," "jerking you around," "stifling discussion," or "a sophomoric attempt to appear clever."


So e.g. the verb, "bite", has several senses including

To cut off a piece by clamping the teeth.

as well as

To hold something by clamping one's teeth.


To attack with the teeth.

We don't, or rarely, need to plump for a particular sense and then define its terms etc. in order to be understood, it's irrelevant. Especially in the context of a discussion (an event in which I was bitten) or argument (the dog is "vicious").


If the other person is asking for the definitions of what you perceive to be "common terms," they might not actually be common terms. Assembling a list of terms which you believe to be "common" is making a bunch of assumptions about how the other person thinks (namely, assuming that you both have the exact same definition of every term). If some of these assumptions are wrong, then the argument will eventually hit a wall where each party is implicitly disagreeing with the other's definition of a contentious term.

The idea that everyone has the same list of "obvious" terms that need no definition is constantly exploited in political discourse. The more common method is for a public figure to make a statement that is "obvious" only to a certain group (namely, the group of people who share a given list). When people outside that group disagree, it is seen as a sign of a fundamental illogic or even stupidity of the opposition, because, according to the in-group's definitions, it's logically impossible to disagree! This is a common tactic to strengthen support among an ideological in-group.

The less common, but more insidious way to exploit these assumptions is to purposely leave some important terms in an argument undefined. When an objection is made, you can then implicitly use whatever definition of the term is best to address that particular objection. To address a different objection, you can simply implicitly change the definition of the term. Even though it's obviously logically inconsistent, the implicitness of this process (you never explicitly say that you're changing your definition) makes it very hard to spot if done well. One of the best defenses against this bad-faith tactic is to explicitly ask for definitions, even of terms that may not be central to the argument. After all, those might be the precise words whose definitions keep implicitly changing.


Seems to me "common" terms are agreed upon terms as defined in reputable dictionaries. A good dictionary offers origin, uses, alternate definitions (synonyms).

IMO, if a person has not had enough exposure to agreed upon terms they should carry a dictionary rather than wasting someone else's time.

It is perfectly proper to ask for clarification when there are possible different interpretations which may have different specific meanings.

The word Potential is such a word. It has several very specific meanings, but the underlying "common denominator" is defned as follows; potential = a latent ability which may become reality.

But if you look up the term potential there is list of different applications of the term.


You must log in to answer this question.

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