Nota bene: English isn't my native language, so when I say acts of the human mind, I attempt to generalize things such as making assumptions, drawing conclusions and (to some extent) to reject.

To me it seems this isn't identical to anthropomorphism, but close to it, or a subset of it. In the Swedish Wikipedia article on anthropomorphism they make a distinction between an entity looking like a human being ("physical anthropomorphism") and having a "human-like spiritual life" ("psychological anthropomorphism"). The latter comes even closer, but to me, it still isn't quite there because of the ambiguity in the notion of someone or something having a "spiritual life".

An example: in one of my school courses, students were criticized by the teacher for including statements such as

Ultimately, our article draws the conclusion that X and Y is Z.

in their articles; the error being that such a statement communicates that the article itself is capable of drawing conclusions as if it were human, or otherwise capable of thinking and reasoning.

Another example: sometimes when writing computer programs, I write things like

the function assumes that the argument n is an integer

in the comments; with this I'm implying (to myself, at any rate!) that if you call the function with a value for n that is anything but an integer, any return values are "bogus" and/or the program stops working correctly. But can a function assume things? In my opinion, not really. On a different note, if anyone else needs to read my code, I'd have to at least further specify what happens when n isn't an integer, but more importantly (to the subject at hand) choose a more proper wording than "the function assumes this and that".

The Swedish noun själsliv (a compound of själ and liv) used in the aforementioned Wikipedia article translates to "spiritual life" according to Google Translate, but a more literal translation is "soul-life", or perhaps "mind-life", and with the latter I think it's reasonable to call this "psychological anthropomorphism".

I notice there's some overlap with Can a non-human “reason” in the sense of inferring. One of the examples from that question entry is

The folk singer's soothing music breathed peacefulness into the cafe.

Now, the music itself can't breathe, obviously, but few English speakers would not interpret this as a metaphor. But when it comes to functions and articles assuming things or drawing conclusions, is it ever "OK", even in an informal sense?

  • 1
    Abbe, I think this question deserves a great +1, congratulations. Commented Jul 26, 2015 at 18:24
  • 3
    I would actually argue that your function really does (literally) assume the argument to be an integer. It expects an integer (another case). Unlike articles, which are utterly static and cannot in themselves act in any way, code is an interactive concept bordering on AI and sentience. The code within your function that causes it to spit out an error or blow up your grandmother (whatever it does when you feed it a non-integer) is really the computational equivalent of creating an expectation, so I don’t think you’re even anthropomorphising your function at all. Commented Jul 26, 2015 at 19:19
  • 1
    @AbbeAbyss Yes (perfectly valid and no anthropomorphising), and no (no subtle difference). I meant “another case” as “an additional case (similar to assumes)”, not “a different case”. Commented Jul 26, 2015 at 19:50
  • 1
    I posit that the function does not assume the argument n is an integer. The implementation of the function requires n is an integer. In source comments, "the implementation" is implicit, one may abbreviate to "the function requires n is an integer". In external documentation, one should differentiate between the domain of the implemented function and the domain of the (Platonic) ideal function that the implementation models. (Somehow, the above seems out of scope for English.SE.) Commented Jul 27, 2015 at 1:26
  • 1
    @JoeBlow (a) Thanks! (b) I did glance; perhaps not in a dictionary but on Wikipedia. I did some "googling" too. I can't recall whether I looked at m-w.com or not but I think I did. At any rate, I still wasn't sure if this is a proper term for the phenomenon I had in mind. (c) I think that my question is long out of necessity, since I felt that I needed to give some examples as to what I had in mind -- describing it more concisely, and/or in purely general terms, would've been difficult for me. Commented Jul 27, 2015 at 10:30

8 Answers 8


Anthropomorphism is not limited to physical appearance, and does not imply specifically a spiritual or soul element.

The attribution of human characteristics or behavior to a god, animal, or object

(Oxford Dictionaries)

So I believe it is the right term.

  • 3
    And in fact, in common usage, it's almost always used in the psychological sense: we talk about anthropomorphising laptops and programmed software and even the climate around us, and none of those are meant In the sense of ascribing the human physique to them. Funnily enough, the word's components taken literally are anthropo: "humanlike", and morphize: "(physical) shaping" - but those origins don't define their actual meaning in practice.
    – Sundar R
    Commented Jul 27, 2015 at 6:35
  • 2
    This answer, and comment, utterly and fully addresses this confusing, over-long question. I particularly admire the conciseness of the answer and comment here.
    – Fattie
    Commented Jul 27, 2015 at 9:25

An alternative to "anthropomorphism" is the word personification. It is almost identical in meaning to anthropomorphism, although perhaps a bit less technical.

According to Literary Devices, personification is:

Personification is a figure of speech in which a thing, an idea or an animal is given human attributes. The non-human objects are portrayed in such a way that we feel they have the ability to act like human beings. For example, when we say, “The sky weeps” we are giving the sky the ability to cry, which is a human quality. Thus, we can say that the sky has been personified in the given sentence.

On entirely another track, you might consider the word metonymy, where one word stands in for another related word, and is frequently used as a device for impersonalising a relationship or object. Thus, the "function" stands in for "the intent of the programmer", the "article" for the "written word and arguments of the author", and so on.

Per Your Dictionary:

Metonymies are frequently used in literature and in everyday speech. A metonymy is a word or phrase that is used to stand in for another word. Sometimes a metonymy is chosen because it is a well-known characteristic of the word.

One famous example of metonymy is the saying, "The pen is mightier than the sword," which originally came from Edward Bulwer Lytton's play Richelieu. This sentence has two examples of metonymy:

The "pen" stands in for "the written word." The "sword" stands in for "military aggression and force."


You ask a lot in this question and the two answers at the time of writing address different things and I agree with them both (anthropomorphism would be an apt term for it). Neither addresses your final question:

But when it comes to functions and articles assuming things or drawing conclusions, is it ever "OK", even in an informal sense?

Yes it is. I would go so far as to say it is "OK" in the vast majority of situations. In general, even in formal and technical settings, but especially elsewhere, it would be fine and has been done in the greatest scientific and technical writings. Most if not all of the greatest works of literature and science would be guilty of it if it were a crime.

To quote Albert Einstein for example:

Current kinematics tacitly assumes that the lengths determined by these two operations are precisely equal

(from On the Electrodynamics of Moving Bodies (aka Special Relativity))

To quote Isaac Newton:

This proposition assumes that P is a relation between relations.

(from Principia Mathematica Volume I)

So we have a theory and a proposition assuming things in two of the greatest works of science.

Having said that, in both those publications when the author is referring to an assumption he is making then the phrase used is "I assume", "we assume" or "it is assumed". They only refer to things making assumptions when it is not themselves doing it.

  • Yeah, I realized that this really is two questions ("what's it called and when is it appropriate?") and changed the title accordingly. Commented Jul 26, 2015 at 17:11
  • 1
    @AbbeAbyss , a useful (and short) page on inappropriate use is en.wikipedia.org/wiki/Pathetic_fallacy
    – Hugh
    Commented Jul 27, 2015 at 3:45

Animism would apply to situations where an inanimate object is believed to possess will, choice, and motive. But that's hardly the right word for the linguistic practice you cited, where that which is authored by a human being (a paper, a function, a work of plastic art, music) is made the subject. We treat the created work as a proxy for the author, as the means by which the author communicates.


I appreciate your differentiation between metaphorical language, which few would question, and non-metaphorical writing. Two things are going on here.

  1. A teacher might question "Ultimately, our article draws the conclusion that X and Y is Z," because indeed, the article is not drawing the conclusion, the authors are. In a course in which the teacher's at least secondary goal is to improve writing, she would want students to note that the sentence will be strengthened with a more appropriate subject: "We (or the authors/researchers) drew the conclusion . . . ."

  2. But were this sentence submitted as part of an article for publication, it could be deemed correct based on the editorial criteria. The editors could frown on the use of I/We and prefer the more abstract expression that would have an article (or more likely, the study) drawing conclusions. This is not the case in my field, but could still be the case in others.


I believe the quotes you list are not merely anthropomorphism (aka personification/prosopopoeia). There's something else there, additional kinds of figure of speech: I think that in the language of rhetoric, these are transpositional rhetorical operations.

Synecdoche is using a part of a thing to refer to the whole; such as referring to a singer by speaking only of his voice.

Metonymy is speaking of a thing by referring to a related thing; such as speaking of the function, rather than the designer of the function.

Figures of speech are perfectly valid in informal speech, but beware of using them in more formal writing, where you should aim to be more literal. At least until your papers are being graded by someone who doesn't care so much about it, at least!


Among other things it is quite common and normal to associate the verb "want" with inanimate objects. For instance, one might say "The car wants to pull to the left," or "I tried typing 'glibnix' but every time I do it the computer wants to replace it with 'glibness'."

Some folks insist that such statements are invalid syntax/semantics because "car" and "computer" are somehow incapable of associating with verbs denoting conscious action, but this is a purely artificial restriction, and does not reflect how the human brain constructs and expresses thoughts.

  • Sure. Considering the statement "The car wants to pull to the left", I think that (depending on context, of course) even to a person that think its syntax/semantics are invalid, there's mutual understanding between the one making the statement and that other person. Also, sometimes (mostly in everyday speech) it'd be cumbersome to communicate such a thing if you're "not allowed to" make this kind of association. (Consider, for example, "The mechanics of the car is in a faulty condition, and this condition causes the car to move to the left when it should not".) Commented Jul 26, 2015 at 19:23
  • 1
    @AbbeAbyss: In a lot of cases, like the car example, you can just replace "wants to" with "tends to". There are other cases where assigning desire/motive to something makes the phrasing easier. And yes, everyone knows what's meant, even those that don't like that usage. Commented Jul 28, 2015 at 6:33

the answer is simply


as Channel, and in a comment Sundar, have explained perfectly.

Your misconception that anthropomorphism does not mean, what it means - is very simply a misconception.

You have exactly the correct word, for the absolutely precise detailed sense you discuss. "It's that simple."

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.