The phrase tries to avoid the overt circular definition found in the variant, "a woman is anyone who identifies as a woman", by swapping woman with female in gender. But is that still a fallacy?

My question is only about the perceived fallacy of the phrase, "a woman is anyone who identifies as a woman," and assuming it is a fallacy, whether any modifications to the phrase (e.g. the one found in the title) make the statement valid.

  • Comments have been moved to chat; please do not continue the discussion here. Before posting a comment below this one, please review the purposes of comments. Comments that do not request clarification or suggest improvements usually belong as an answer, on Philosophy Meta, or in Philosophy Chat. Comments continuing discussion may be removed.
    – Philip Klöcking
    Commented Apr 5, 2023 at 17:16
  • This entire question is basically just the title -- where the short question-body is just a note about the title. As such, it's basically just, "Is [some statement] a fallacy?", which, without further clarification, is essentially asking, "Is there a way to interpret these words such that they don't invoke a fallacy?".
    – Nat
    Commented Apr 7, 2023 at 8:25
  • 1
    As answers have pointed out, this sounds like a trivial question -- because, yes, there exists a non-fallacious interpretation of those words, e.g. that they refer to a definition. But is that really what you meant to ask? If so, it might help to edit this question to clarify that, yes, all you wanted to know is that there's a non-fallacious interpretation of those words. Alternatively, if you did mean to ask about if some specific concept is fallacious, then you may want to edit the question to clarify the concept.
    – Nat
    Commented Apr 7, 2023 at 8:31

8 Answers 8


A woman is an adult that identifies as female in gender.

A fallacy is an argument that is specious but persuasive. You have presented no substantial argument which often takes the form of first premise, second premise, conclusion. This language should probably be adjudged as a real definition, which is a summary of analysis which seeks to establish a relationship between words and characterize properties of real things. The terms in question in this topic are usually, 'woman', 'gender', and '(biological) sex'. In particular, the claim also invokes 'identify', which means that whether or not this definition is acceptable logically presumes certain semantic relationships among all four terms which include sexual, but much more importantly, psychological dimensions. Gender is a belief about how one's experience of the world and self-image aligns with one's biological inheritance.

While I'm not a philosopher of sexuality (SEP), anecdotally, the arguments I have seen roughly state this:

  • A 'woman' is a term for a type of person.
  • A person can have a 'gender' and a 'biological sex' that do not align according to a crisp, binary dichotomy of traditional male-female concepts such as 'intersex individuals' who may have reproductive organ systems of both 'biological men' and 'biological women' for instance.
  • Even if an individual is either a 'biological male' or 'biological female' but doesn't have both reproductive systems, there is a psychological aspect to sexual identity in which one can 'identify' with various thoughts and impulses. For instance, a 'bisexual' or 'homosexual' person has 'sexual impulses' towards the same sex.
  • Some people have atypical 'sexual impulses' that lie on a spectrum such as 'asexuality'.
  • Much more important than 'sexual orientation', it is possible to have a 'biological male' who 'identifies' as a 'psychological female' not just in 'sexual impulses' towards men which might be 'homosexuality', but also in psychological ways that might be construed as 'womanhood' such as wanting to have female sexual organs, want to do activities that are typically construed as female in society, wanting relationship with women as a woman and not a man. This notion of 'womanhood' is often called 'female gender' as to distinguish the psychological and biological aspects of 'identity'.
  • Such a person then might transition their 'biological sexuality' to conform with their 'psychological identity' which is often seen as more of construing 'personhood' than mere 'sexual orientation'.

According to the NIH's definition on gender:

Gender refers to socially constructed roles, behaviors, activities, and/or attributes that a given society associates with being a woman, man, girl, or boy, as well as relationships with each other. As a social construct, gender varies from society to society and can change over time.

Ultimately, "roles, behavior, and activities" is far broader than sexuality and biological sex. Emphasis on gender as distinct from sexuality and biological sex means that while sexuality may play a role in identity, that is, gender identity for many subsumes sexuality in a small way, the emphasis is often on the non-sexual aspect because skeptics of non-traditional gender identity attempt to conflate sexuality and gender to claim gender doesn't exist in it's own right. This is important in cultures that strictly prescribe and proscribe 'manly' and 'womanly' thoughts, behaviors, roles, etc. assigned to one's biological sex by society. For a misgendered person, this can be a source of extreme friction with society and lead to mental health issues ultimately, in the most severe cases, culminating in suicidal ideation and suicide.

Critics of this thinking (besides moral absolutists who point to religious doctrine and just utter "sin!" or "bad!" or "boo!") often try to construct an argument that such thinking offered above is some form of gender dysphoria. Such arguments, as I have seen them anecdotally, are often unsophisticated and don't understand the four D's used in abnormal psychology as a basis for justifying the application of the term 'mental illness'.

So, philosophically, what we have going on here is an argument between two sets of people who relatively affirm or disconfirm the statement you have presented that are looking to see the "illogic" in the other side's position. That back and forth is often colored by the degree of sophistication in terms of the processes of definition, logic, and argumentation. As such, the statement you provided is perfectly reasonable with a very flexible, nuanced vocabulary and not within a more rigid, simpler vocabulary. But on the face of it, this statement is not really so much an argument but a definition whose acceptance hinges on a much deeper analysis based on a worldview.


"A woman is anyone who identifies as a woman" is a definition, not an argument (it defines what "woman" means). So it cannot be fallacious. But it's circular*, which means it's not a very useful definition. Definitions are supposed to tell us what things mean, but this doesn't do that.

As an analogy, consider: "a flarposjprpahopg is anyone who identifies as a flarposjprpahopg". This provides no insight into what "flarposjprpahopg" means, other than it's something that it's someone can identify as.

*: One could argue, technically, it's not circular, because it's possible for someone to identify as a "woman" or a "flarposjprpahopg" regardless of what those words mean. But for it to make sense for someone to identify as such, one would need to know what the term means, which requires a definition. If the definition only entails identifying as such, it would not make sense to identify as such.

Definitions are not objective

If someone takes the above definition (or the definition of "A woman is an adult human female" that is typically used by anti-trans people, with "female" referring to sex), as an objective fact of reality, that would be fallacious, because definitions of words are subjective.*

The definitions we choose to use may have consequences for others, such as whether we are inclusive and considerate towards trans people, in the language we use and in the systems we've built around that language (e.g. gendered bathrooms and changing rooms and prisons), whether they're allowed to access scientifically-verified effective medical care, etc.

And yes, while the definitions we use are subjective, this wouldn't excuse someone choosing to not be a decent human being by using a definition that results in harm to others. On the contrary, it puts a lot more responsibility on each person, to choose a definition that doesn't result in harm, or that results in as little harm as possible.

*: Although one could also state a definition roughly (but not strictly speaking) as a fact, to say that's what the definition should be, to propagate that definition. This would not be fallacious.

Replacing "woman" with "female"

Yes, that could work, although arguably not as you did.

If we assume commonly-used (progressive and scientific) definitions of words, "female in gender" is somewhat nonsensical. "Female" (as a noun) refers to sex, and it doesn't make much sense to say "sex in gender". If you treat "female" as referring to gender (i.e. a woman), you'd be back to a circular definition.

We could, however, use a definition such as: "A woman is someone with an internal sense of being female". This would have woman (gender) reference female (sex), so it would therefore no longer be circular. It also seems to match the relation between gender and sex, as they're typically defined by pro-trans people and scientific organisations (e.g. the WHO).

Additionally, this definition would avoid the common confusion/ambiguity with what it means to "identify as" a woman. Pro-trans people typically mean this as having this internal sense of being female (and saying you're a woman is just communicating this internal sense to others), whereas anti-trans people typically take this to mean someone simply needs to say they are a woman on one occasion.

Note that having an "internal sense of being female" does not mean someone believes themselves to be biologically female nor to have genitalia that they don't have, but rather that they have a female brain, roughly speaking, along with the associated psychological traits and characteristics.


It isn't an argument, so cannot be a fallacy.

It is merely a definition.

A definition can be useful or not. It can be a prescriptive definition or a descriptive definition. Prescriptive definitions are common in maths "A group is a set of objects with certain properties." This is prescriptive. It means that "in this work the word 'group' has this definition." A descriptive definition is a dictionary definition, it attempts to capture the definition of a word as used by speakers of the language. A dictionary definition is useful if it accurately captures how the word is used by speakers of the language.

This isn't a circular definition. A circular definition would be "A woman is a woman" or at one remove "A woman is an adult female human, and female is the adjective that is used to describe women".

This is not circular. There is a class of adult humans that identify as "female". This class of humans could be given a name. That name could be "women".

You can make a prescriptive definition however you want - "In this paper I shall use the term 'woman' to denote a pot with a handle, spout, and lid, in which tea is brewed." It is confusing to use a common word with a technical meaning like that but mathematicians do it all the time. You could, prescriptively, choose to define "women" as "adult humans that identify as female". But you should be aware that your choice does not imply that everyone else has to use the same definition. It is a statement of what you feel is a useful definition in some context.

You can judge a descriptive definition on whether it captures the use of the word by other speakers. You can say that a definition is wrong when it doesn't match the use of the word. If the dictionary defined "woman" as "a pot with a handle, spout, and lid, in which tea is brewed" you would say it is wrong, because demonstrably other speakers do not use the word "woman" for that.

You can criticise a dictionary defintion because it doesn't capture the way that you personally use a word. That's allowed because you are a speaker of the language and so have a right to judge how well a defintion captures your use a word. But others may not share your opinion.

Word in natural language are "fuzzy", have multiple senses and are used differently by different people. There may be a practical purpose in clarifying the definition for a particular context. Whether your definition is the right one depends on what purpose it serves. If a particular definition is required for a particular context we can craft one.

So, not circular, not a fallacy, just a matter of opinion.

Now I'm going to find a woman and make a cup of tea, if you know what I mean.

  • 5
    And indeed, most of the words we use to describe classes of things are actually fuzzy around the edges. Nouns like "disease" or "influencer" or "road" or "restaurant" all leave room for doubt about exactly what falls into the category. Even with legal definitions, it's left to the courts to decide whether something is or is not a "contract". Our words are fuzzy and ambiguous, by design; if we need more precision in a particular context (like sending someone to a women's prison), we qualify the word accordingly. Commented Apr 6, 2023 at 0:21

The following definition is not circular:

A woman is somebody who says they are a woman.

This definition proposes a test, "do they say they are a woman?", to determine if somebody is a woman (according to the given definition). This test can be performed without needing to circularly apply the definition of the term "woman" ─ because we don't need a definition of "woman" to know whether or not somebody says they are a woman.

You may argue it is not a useful definition, because it does not depend on what the person who says "I am a woman" means by the word "woman", only that they use that word to describe themselves. Others will disagree. But the definition itself is not circular.

Perhaps it will help to make an analogy with a similar non-circular definition which was used historically, though is no longer used in modern times, but the definition was not contentious and I am not aware of anybody seriously arguing that the definition was invalid due to circularity.

Before marriages had legal status in modern law, it used to be that a husband and wife became married in a ceremony, in which a religious leader declared "I now pronounce you husband and wife". This pronouncement itself used to be what made two people husband and wife, so if two people had not been married in such a ceremony where such a pronouncement was made, they would not be husband and wife.

So the definition of "husband" and "wife" included that the husband and wife had been pronounced as such, by the power vested in whoever officiated wedding ceremonies. (There were other aspects to the definition as well, but this criterion was required.) Does this mean that until modern times, marriages were meaningless, because being a "husband" or "wife" depended on a pronouncement being made, where the pronouncement itself necessarily included those terms which were defined by the pronouncement?

Of course not. This definition is likewise not circular, because we can apply the definition to determine if two people are husband and wife ─ i.e. has such a pronouncement been made by someone qualified to make it? ─ without having to more deeply investigate the meaning of the words in that pronouncement. The fact of the pronouncement being made, regardless of its meaning, is enough to satisfy the definition.


In Logic, a fallacy is somehing that is provably false. In order to make such a proof one has to agree on the initial values.

Here the problem is one of context (as is usual with natural language). In what context is this statement being made? Without knowing, one can't begin to start a sensible argument about whether it's true, false, or in need of further clarification.

It does, however, reek of circular definition. It's very easy to reduce this to "A woman is an adult who believes that she is a woman". Which might be be what is intended, but is clearly false in the context of a cervical cancer screening program. A woman in this context, is a person who possesses a cervix. A biological male whose genitalia developed normally, does not have this anatomical detail, and identifying as female cannot change that. A female-to-male transsexual, in contrast, does have a cervix, at least before surgical modification of his body.

(For avoidance of doubt don't assume I am unsympathetic to the plight of a person whose brain developed with female characteristics, trapped in a male body).



This definition is logically self-consistent and non-circular. It matches how some native speakers of English use the term, but not others.

One reason many people prefer it is that “biological sex” turns out to be impossible to define without being arbitrary or declaring some people with intersex conditions to be exceptions. You could get around this by abandoning the idea that there are only two genders, and introducing some third category like “intersex,” but then it would be necessary to define that.

As a thought experiment, imagine that Alice and Bob have had a brain transplant. Alice says, “I’m still a woman, just in a man’s body. In fact, you can confirm this biologically: if you do a biopsy of my brain, you will see that it has XX chromosomes. There are biological differences between myself and a man; but they’re not externally visible.” There’s no logical fallacy there. While there’s no one theory that everyone who uses people’s preferred pronouns agrees with, many transgender activists (for example, Julia Serano) argue that the situation with trans people is like this: they are biologically the gender they say they are, but the biological difference between them and cis people is located in their brains and difficult to directly observe. They also claim that no other causes of genuine gender dysphoria exist. This might or might not be factually sound, but it is not logically fallacious or circular.

If physicians made a list of “male” and “female” traits, some of the items on the list might be:

  • XX versus XY chromosomes
  • Having breasts or not having breasts
  • External genitalia
  • Attraction to men or women
  • Levels of testosterone and estrogen
  • Being able to get pregnant, or impregnate others
  • Being taller or shorter
  • Having a ring finger the same length as the index finger, or longer
  • A biological sense of gender identity in the brain

We know that the last one belongs there because, up until the 1970s, many doctors incorrectly thought that there were no differences between male and female brains, and recommended that babies born with ambiguous genitals have them surgically altered and be raised as the gender their parents picked for them. In fact, when this did not match their internal gender identity, those kids grew up knowing they were the other gender than their parents told them they were, and developed gender dysphoria.

And, if you go down the list, for every definition of “biological sex” on that list (or that I’ve ever heard proposed) there is some intersex condition that breaks it. There are people with XY chromosomes and women’s bodies (congenital androgen insensitivity syndrome), men with gynecomastia, people whose chromosomes are XXY, XYY, or XO, people born with a vagina who later grow a penis, and even hermaphrodites who got each other pregnant.

So, if we want there to be only two genders defined by “biology,” the only way to make that work is to arbitrarily pick one item on the list and say, that is the one trait that defines biological gender. We can’t pick two or more, without needing to invent another category for people who have ine but not the others, or something that’s neither. And if we’re going to resolve this by picking a single defining trait of gender identity, there’s a strong philosophical argument that we should pick one in the brain.

  • You are committing the grey fallacy. That extremely rare ambiguous cases exist (e.g. intersex conditions) does not mean that the same ambiguity extends to everyone, and that anyone can be a member of either biological sex. That's a complete non sequitur.
    – user76284
    Commented Dec 1, 2023 at 6:20
  • You are trying to muddy the waters by appealing to intersex conditions, which have nothing to do with transgenderism. That there's ambiguity in the former does not imply there's ambiguity in the latter, and that therefore anyone can identify as whatever they want (man, woman, cat, dog, horse, alien, etc.)
    – user76284
    Commented Dec 3, 2023 at 21:47
  • @user76284 That is still not a correct summary of any argument I made. Please re-read my post.
    – Davislor
    Commented Dec 4, 2023 at 1:26

Unlike other answers, I will follow the Cambridge Dictionary definition of "fallacy":

an idea that a lot of people think is true but is in fact false

Following that meaning, a statement like this very well can be a fallacy. However, to establish whether or not it is one we would have to be able to establish the two facts required:

  1. a lof of people think it is true
  2. it is in fact false

That this question has been marked as "controversial" is a good indicator that we do not have an established consensus on #2. The discussion on this question is ongoing. We cannot say whether this statement is true or not. Maybe in the future a consensus will emerge and we can state that - for that time and culture - the statement is either true or false. As of now, any such claim would simply be a part of the ongoing discourse.

That said about the general case, in some specific contexts the "who identifies as" part has been made a rule, either in law or in bylaws. In these specific contexts, the statement has been codified as true within a particular domain. This leads to the case of Avi Silverberg, which under these very particular circumstances might be an example of both 1 (the rules of the competition say so and the people within this domain agreed to these rules) and 2 (Mr. Silverberg is not a woman under any definition) being true. (The whole thing is widely believed to have been a protest action.)

In general, at this time our society has an ongoing discussion about the meaning of words such as "woman", "man", "female", "male" and even "identify". With most of the words that make up the statement being under review, we cannot, at this time and within western culture, determine whether or not this statement is true, false, a fallacy or not.


It's not a fallacy, because it's not an argument. But the actual problem with these definitions is that they aren't terribly useful because they kinda miss the point.

The thing is you can interpret a question like "what does it mean to be X" from at least 2 angles, one is as a demarcation line between X and not X and the other is as a description of what X have in common what socio-economical-cultural-... consequences arise for an individual because of being X. What is the X condition of life?

And while both your definitions kinda fulfill the requirement of a demarcation line. Whoever self-identifies as group X or with properties X/Y/Z is a member of group X. That in isolation is completely meaningless.

Like the actual consequence of what it means to be a woman based on that definition is to say that you are a woman, nothing more and nothing less. It's just a name. But it is not just a name, neither for the groups that currently is described by the label nor by the group that wants to limit that description and neither by the group that wants to expand that description.

They all more or less take for granted that "being a woman" has several consequences, like being addressed with female pronouns for example. So as the consequences are linked to the group, they can also serve as (part of the) definition of the group. So in addition to woman being a label of the self, woman is also a group that ought to be addressed with female pronouns.

And from those consequences of being a woman and the attempt to achieve or deny them to certain individuals or groups, arises the struggle about a demarcation line of what it means to be a woman (under the hidden assumption that the consequences will remain the same). So to talk about self-identification as a demarcation line of what constitutes membership of the group "women" in isolation misses the point entirely, because it's unclear what assumption that makes about the consequences and whether it ends up being a fallacy, an opinion or dispute.

Because it only can be a fallacy if you take certain premises as fixed and draw conclusions from them, but that's actually more of an attempt to demand or to negotiate such a premise which in that sense can't be fallacious because it's not an argument.

And just in general you've got the problem that what constitutes being a male/female/man/woman is manifold. A biologist might distinguish based on gametes (sperms and egg producers), (though also not exclusively and that also doesn't always work as expected), an engineer distinguishes between sockets and plugs, as most people wear clothes and don't carry high resolution microscopes with them, the distinction between gametes and via primary sexual organs isn't possible and so they go by secondary sexual characteristics. But even those might fail you when they aren't pronounced or if they used to be there but aren't. Like if a man with testicular cancer gets their testicles removed they still used to be a man, are visible as man and probably identify themselves as man. So what we actually might end up basing that distinction upon can also include completely non-biological factors like, choice of clothing, makeup, mannerisms, experiences, upbringing, socialization, self-identification and so on and on and on.

And unless you look at a very specific domain where you can limit the number of factors that are relevant to the distinction, you'd end up with a combination of all these features and more and the weighting of these features might not even be consistent from person to person. And while some of them can be a binary others are better described by a spectrum.

So if and where a demarcation line makes sense depends on the very specific domain that you are interested in and if you want a universal line that is always "true", then this more or less boils down to a political struggle over where to draw such a line and is not really a matter of logic.

But regardless of where you draw the line what people care about is usually not the line itself but the consequences that arise from being on the wrong side of that line.

  • 5
    Actually a biologist would probably say that gender and sex are distinct, sex is defined by gametes, but that isn't always correct and also it's complicated (on a purely biological level), and gender is in the brain. Because that's what the science points to. Here's a biologist saying exactly that (there's also a massive list of references at the end of that video).
    – NotThatGuy
    Commented Apr 5, 2023 at 23:36
  • @NotThatGuy You're right that was a gross oversimplification and scientific domains like biology are also evolving their views and definitions with the facts. Also thanks for the video!
    – haxor789
    Commented Apr 6, 2023 at 6:47

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