Post-nominal letters

[Asher Smith]

Based on some conversation in the Back to Methuselah thread, I would appreciate a ruling from Alex that I can pull together a PR for to merge into SEMoS and then start normalising across the corpus.

SEMoS 8.10.5 currently applies to academic degrees, but can be broadened to cover all post-nominal letters. Here are some questions:

• The guidance currently says to put periods in the middle of post-nominal letters, and gives an example of D.D.S with no period at the end. It seems like the CMOS suggests that if you're using periods (which as far as I can tell CMOS18 does not suggest at all; in my experience with modern news media they generally don't), you should have them after all words, and I cannot find any other examples in the corpus of no period at the end of a string of three or more post-nominal letters. What's the source of this rule? Should it apply to other post-nominal letters?
• Some common post-nominal letters in the corpus are F.R.S. and O.B.E., and they are an absolute mess of <abbr epub:type="z3998:name-title"> and <abbr epub:type="z3998:initialism"> tags. Should the correct formatting include both, like <abbr epub:type="z3998:name-title z3998:initialism">? If so, does the order of the two tags matter? (And generally speaking, when something should get two tags, does the order matter?)
• As an aside, PhD is written a bunch of ways in the corpus; none of them have a period in them, though the SEMoS says that D.Ph. should.

Cheers!

[Vince]

I’m assuming the D.D.S is just a mistake in the manual. It happens. SEMoS goes its own way on a variety of things, including this one. CMS recommends no periods for academic degrees, unless it fits “a journal’s established style.” This is our “established style.” :)

A D(r)Ph and a PhD are not the same thing. I’ve only ever seen a DPh as DrPh, though. SEMoS should probably specify what to do with PhD; it is far more common than D(r)Ph. (Should we include the hairspace between the Ph and the D like the opposite? Should we include initials? Etc.)

SEMoS 8.10.5 answers the last question—if they have name-title, they don’t get initialism. (Clean sorts class names in alpha order except eoc, but I don’t believe we do that for epub:types.)

Although “absolute mess” is a bit strong, it is quite common to see multiple things in the corpus for a given rule. First, rules change over time, and often the corpus is not updated when that happens if it’s too burdensome or there’s no one to do it. More importantly, everyone involved here is human, and we miss things, as producers, as reviewers, as the EiC. This is why I repeat, it feels like weekly, that if SEMoS says do X, it’s irrelevant that Y and Z exist in the corpus: X is what we do.

It should (might?) be possible for lint to search for a known set of degrees and other related abbreviations and determine whether they had the correct epub:type. As always, it just requires someone to do it.

[David]

Rabbit hole alert: I've never (to my recollection) run across a "DPh", but rather "DPhil", most famously Oxford, but also a couple other UK universities, and occasionally German (e.g. Bonn). Not that this matters materially for Asher's query! And in this case, DPhil and PhD are equivalents, just different nomenclature.

So what's "DPh" meant to be, then??

David / Fife, UK

On Sunday, 22 February 2026 at 01:06:39 UTC Vince wrote:

I’m assuming the D.D.S is just a mistake in the manual. It happens. SEMoS goes its own way on a variety of things, including this one. CMS recommends no periods for academic degrees, unless it fits “a journal’s established style.” This is our “established style.” :)

A D(r)Ph and a PhD are not the same thing. I’ve only ever seen a DPh as DrPh, though. SEMoS should probably specify what to do with PhD; it is far more common than D(r)Ph. (Should we include the hairspace between the Ph and the D like the opposite? Should we include initials? Etc.)

SEMoS 8.10.5 answers the last question—if they have name-title, they don’t get initialism. (Clean sorts class names in alpha order except eoc, but I don’t believe we do that for epub:types.)

Although “absolute mess” is a bit strong, it is quite common to see multiple things in the corpus for a given rule. First, rules change over time, and often the corpus is not updated when that happens if it’s too burdensome or there’s no one to do it. More importantly, everyone involved here is human, and we miss things, as producers, as reviewers, as the EiC. This is why I repeat, it feels like weekly, that if SEMoS says do X, it’s irrelevant that Y and Z exist in the corpus: X is what we do.

It should (might?) be possible for lint to search for a known set of degrees and other related abbreviations and determine whether they had the correct epub:type. As always, it just requires someone to do it.

[Vince]

Doctor of Public Health. It’s possible the SEMoS example is short for a DPhil, but like DrPh, I’ve personally only ever seen DPhil abbreviated as … DPhil. :) (Wiki does indicate that sometimes PhD is Ph.D. in North America.) So, really, D.Ph. is ambiguous; I’m not sure which it’s for.