definition of `<mentioned>` too restrictive

[sydb]

Ash Clark came across the following passage in An Essay to Revive the Ancient Education of Gentlewomen, a 1673 work by Bathusa Makin.

I confeſs this part of Speech is moſt difficult to be known in the Engliſh Tongue; yet it may be done thus, All words ending in ing, d, t, or n, which have no ſign at all, and may be reſolved into Verbs, are Participles, as learning, which doth learn; learned, which is learn­ ed.

It is clear (at least to me) that the “ing”, ‘d’, ‘t’, and ‘n’ are mentioned, not used. The description of <mentioned> in the tagdoc (and thus, in some sense, the definition of that element) is “marks words or phrases mentioned, not used”. Even the more complete description in the prose says <mentioned> is for “where a word or phrase is being discussed in the body of a text rather than forming part of the text directly”. But these things are neither phrases nor even words; they are most certainly sub-word thingies.

Of course a word can be mentioned, not used. (My father taught me this discrepancy ½ a century ago with something like “Boston is a city of .6 million people; ‘Boston’ is six letters long.”)

Certainly a phrase can be mentioned, not used — “The phrase ‘man’s best friend’ is often attributed to Frederick the Great of Prussia.” But so can a clause — “The saying ‘Injustice anywhere is a threat to justice everywhere’, extracted from King’s Letter from Birmingham Jail, often appears on lawn signs and bumper stickers.” (Yes, one might choose to encode either of those with <q> or <quote>, but one might also quite reasonably choose <mentioned>.)

And, as shown above, so might a suffix or prefix — “The ‘epi’ of ‘epinephrine’ is analogous to the ‘ad’ of ‘adrenaline’.”; or, for that matter, a lemma or stem — “The ‘nephr’ of ‘epinephrine’ is analogous to the ‘renal’ of ‘adrenaline’.” Or even a letter — “An ‘a’ can appear drastically different in different typefaces.”

At the moment, I am not entirely sure what (if anything) should be done about this.

[bansp]

While not beautiful, the following might do the job, perhaps:

• “marks words or phrases mentioned, not used” --> "marks textual fragments mentioned, not used"
• “where a word or phrase is being discussed in the body of a text rather than forming part of the text directly” --> “where a fragment (most typically a word or a phrase) is being discussed in the body of a text rather than forming part of the text directly”

It only implies other types of spans rather than enumerate them, but then, the examples you've come up with would round the spec up.

[lawrenceevalyn]

Is 'textual fragments' overcomplicating it? What about "marks text that is mentioned, not used" ? and "where text is being discussed in the body of a work rather than forming part of the work directly"?

[ebeshero]

I like @lawrenceevalyn's concise solution. As I was reading this I was thinking, "particles" or "grams," but we need not that complexity. :-)

[sydb]

Indeed. Both @bansp’s and @lawrenceevalyn’s solutions had me thinking “why didn’t I think of that?” :smile:

[martindholmes]

How about "words, phrases, or components of words"?