Students of linguistics, literature and philosophy sooner or later get the idea that there is an important distinction between the use of a word (any word) and its mention. In print, the distinction can be marked by putting on punctuation marks to indicate that a word is being mentioned (quoted); alternatively, it can be italicised. The common purpose is to prevent confusion about who said what. It also, in some circumstances at least, exempts someone who mentions a word from any opprobium which might attach to it use. But it’s not always so simple.
A police officer is giving evidence in the kind of ordinary court of law found in many countries, and says at some point, “I then arrested him and he used a racist epithet”. Now because the police officer is giving evidence and is not prosecutor, judge and jury all at the same time, it is entirely legitimate and relevant to ask, “What epithet?” To know it may be relevant to assessing the gravity of an offence. The officer is only being asked to mention the word, to quote it, not use it themself. It ought to be simple. Sometimes it isn’t and the police officer may demur, “I don’t want to say the word”. In that situation, a judge may ask the officer to write it down, knowing that this is usually acceptable even when saying the word is not. The slip of paper may then be passed silently to judge, prosecution, defence, and jury.
The officer’s hesitation may be prompted by different kinds of sensibility - they may simply not want to be party to circulating the word in any form, use or mention; they would like the word to go away and not saying it is a step in the right direction. Even if the officer does say it, a newspaper reporting the case will most likely not print it. Instead, a report may repeat the original “racist epithet” formula or, alternatively, print the word in a censored form, say, ******, which may be modified by providing one or two letters as clues.
This curious practice of giving clues is a modification of the slip-of-paper compromise: the reader now does not have to see the word, but is enabled to infer it, and the more clues provided the less uncertain becomes the inference until the word is staring you in the face. If you want to check the first line of Philip Larkin’s poem, This Be The Verse, online sources will offer you as the second word several versions: “****”, “***k”, “f**k” and “f*ck”. It is an interesting question why anyone should think “f*ck” preferable as an alternative to what Larkin actually wrote in those heady days back in the 1970s when people were trying to say and write what they meant. The obvious answer is that they now mis-quote it as “f*ck” because they do not even want to see the complete word, just as the police officer did not want to hear the racist epithet even in the form of a mention.
Especially in relation to speech, there may also be a fear that your audience might react in an unwanted way. A police officer who says, “I then arrested him and he used a fat-shaming epithet” may not want to mention the word or expression used simply from anxiety that the court-room audience might not be sufficiently on guard to suppress a titter. It’s possible that they have already had such a thought about the unfortunate officer. An epithet can be well-chosen, even if disgraceful or illegal.
There is a back story relevant to the discussion. In all the main monotheistic religions, use or mention of the name of God is hedged about with prohibitions, taboos, and contextual requirements; it is one of the Ten Commandments that “Thou shalt not take the name of the LORD thy God in vain; for the LORD will not hold him guiltless that taketh his name in vain” (Exodus 20:7). This requires interpretation, and indeed since it was first written down, many (millions of?) man-hours - paid and unpaid - have been devoted to its exegesis. One fairly common interpretation leads to the conclusion that, really, one should not use the name of the **** at all, though whether that is a matter of piety or prudence has also to be decided.
There is also a heresy within Russian Orthodoxy which goes in the reverse direction. The name of God should not be used carelessly because the name of God is God, rather in the way that some mathematicians think that the number names are the numbers, without any other kind of existence than the words we commonly employ - though that existence is reckoned well-worth having.
Those who believe that the name of God is God (the heresy is still alive) are called Imyaslavtsy, meaning Those who glorify the Name. When in 1913 Nicholas the Second of Russia was told that Russian monks on Mont Athos had become infected with the new heresy, he despatched a gunboat and two transport ships to Mont Athos. The Archbishop of Vologda was put ashore and held lengthy talks in which many monks identified themselves as heretics and refused to recant. As a result, initiative was passed back to the repressive apparatus of the Russian state. Troops came ashore, rounded up the heretics - killing four, injuring around fifty - and eventually loaded over eight hundred monks onto the ships for transport to Odessa where a few were found Not Guilty and allowed to return to Athos; rather more were jailed; and the remainder defrocked and sent into internal exile.
This little discussion actually has an ultra-modern relevance. There has always been a part of linguistics, notably that associated with the making of dictionaries, which has interested itself in first uses of a word, subsequent developments in the way the word is used, and in some cases, a word’s fall into disuse. Such corpus linguistics was heavily dependent on printed texts and was extremely laborious work. Modern computer-based data harvesting radically alters the situation: provided it is online, truly huge amounts of data can be harvested with ease. Take any word which, say, has recently become popular and it will be possible to track its origins, its often-global dissemination, its typical users (classified along any dimensions you like), and so on. But there is a catch.
Suppose I want to give an example of a newly popular word and choose “genderfluid”. Then if what I write should appear online, a data-harvesting program designed to pick up occurrences of the word will pick it up. But I haven’t used the word; I have mentioned it. And unless the program is trained to distinguish use and mention it cannot ground certain interpretations which human users of the data might want to make. Unable to distinguish use and mention, a program would not differentiate between such very different occurrences of “genderfluid” as these:
(1) I am genderfluid
(2) I never use the word “genderfluid”; I would not like to take it in vain.