Do footnoting superscripts go inside or outside punctuation?

Question

When using superscripts to indicate a footnote, do these fall inside or outside adjacent punctuation? If there is an answer, is that answer applicable worldwide, or just to specific regions or publishers?

Does it matter what the particular punctuation is, including such punctuation as commas, colons, parentheses and other brackets, periods, and quotation marks?

Does it matter whether the footnote applies to just one gloss¹, or to an entire phrase in toto?

Does the answer change if, instead of using instead of numeric footnotes, you use the traditional sequence of symbols (*, †, ‡, §, ‖, and ¶)²  ?

---

1. OED: “A word inserted between the lines or in the margin as an explanatory equivalent of a foreign or otherwise difficult word in the text; hence applied to a simliar explanatory rendering of a word given in a glossary or dictionary.”

2. As enumerated on pp 68–69 of Robert Bringhurst’s The Elements of Typographic Style (version 3.2); Hartley and Marks, 2008. Bringhurst goes on to say “But beyond the asterisk, dagger³, and double dagger⁴, this order is not familiar to most readers, and never was.”

3. That is, the † character at codepoint U+2020 DAGGER, also known as the obelisk, obelus, or long cross. The classical plural of obelus is obeli.

4. That is, the ‡ character at codepoint U+2021 DOUBLE DAGGER, also known as the
   diesis or double obelisk. The classical plural of diesis is dieses.

Answer 1

A lot depends on what style manual you follow. I follow Hart’s Rules at the University Press Oxford and according to Hart’s -

Footnote references should be placed outside punctuation, but inside the closing parenthesis when referring to matter within parentheses.

It makes no distinction between numeric or symbol footnotes.

It makes no distinction between a single word or phrase.

Footnotes should begin with the numbers indented 1 em space.

Answer 2

In my business, it depends on what the journal publisher says is the rule. Most journals require them to be outside the punctuation. Some, like all the journals published by the Nature group, require them to be inside the punctuation. There are reasonable arguments for whatever style is mandated, but there's no internationally accepted style.

Although it's traditional to place the superscript footnote inside the semicolon and colon but outside the comma and period, some (very few) biomedical publishers don't like that foolish inconsistency: they demand that all superscript footnotes be outside the punctuation.

I've never seen a style manual that says it matters "whether the footnote applies to just one gloss, or to an entire phrase".

In biomedical journals, the traditional symbols aren't allowed in the text, only in the list of author names, and it doesn't matter: some publishers want them inside and others want them outside the punctuation.

All rules are ad hoc, it seems.