Talk:-sis

Other usages?

It seems as though the two listed applications of -sis don't cover all usages. Is "basis" a "noun of action or progress?" A basis seems to be rather inert. But a basis is obviously not a medical condition, either. The same goes for words like "oasis," "nemesis," "thesis," and derivatives of "thesis" like "hypothesis" and "antithesis." And while "diagnosis" and its cognate "prognosis" are used in medicine, they aren't medical conditions, as in you can't suffer from a chronic case of "prognosis." I suppose "diagnosis" would be a "noun of action," though, and "prognosis" is perhaps a "noun of progress." I'm going to be Wiktionary:BOLD in updating this page, but I'm open to feedback! Pdxuser (talk) 22:05, 18 December 2020 (UTC)

RFD discussion: October 2022–January 2023

The following discussion has been moved from Wiktionary:Requests for deletion (permalink).

This discussion is no longer live and is left here as an archive. Please do not modify this conversation, but feel free to discuss its conclusions.

---

-sis

A purported suffix, but I recognized all the members of the associated category as direct borrowings from Latin or Greek. DCDuring (talk) 22:18, 12 October 2022 (UTC)

Comment. If, as has been suggested, we analyze e.g. vitaminosis as vitamin +‎ -o- +‎ -sis, we will have plenty of late formations.  --Lambiam 15:22, 13 October 2022 (UTC)

I consider the vowel in -osis more essential (lemma-like) than the o in -otomy. This is a good place to look at how other dicts do it. There appears to be a small number of words including merisis that were formed in English and don't have the -o-. I've cleared out some bad manual categorization and redirected some links to -osis. Ultimateria (talk) 17:59, 16 October 2022 (UTC)

So keep and convert the page to a rare alternative form of -osis. Ultimateria (talk) 18:03, 16 October 2022 (UTC)

@Dan Polansky Note that your solution of deleting all forms starting with "o" would be highly unsatisfactory in this example. More evidence that there are simply two variants. Theknightwho (talk) 09:44, 21 October 2022 (UTC)

Exceptions can be made on a case-to-case basis, perhaps along the lines of Ultimateria's "I consider the vowel in -osis more essential (lemma-like) than the o in -otomy." There may be a case for keeping -osis: we would need to have a closer look at the etymology and the like. Analyzing -osis as -sis would probably not be wrong, although rather unconventional. Turning all the following redlinks blue still seems undesirable to me (for the reader, a full discussion is at -otomy RFD): -ocyte, -ogenesis, -ogenic, -oleous, -olysis, -ophil, -ophile, -ophilia, -ophilous, -ophyte and -opore. --Dan Polansky (talk) 10:17, 21 October 2022 (UTC)

@Dan Polansky So now you're positing the existence of a linking morpheme -o- and variable stems in different cases? Doesn't sound like Occam's razor to me. The fact that you're worried about creating redlinks only confirms my suspicion that the real reason for your position is still to do with how words get categorised (even if you won't admit it). Theknightwho (talk) 10:22, 21 October 2022 (UTC)

In the medical sense of an abnormal condition, -osis (as in halitosis) is a productive suffix, probably formed by reanalysis of terms like cyano- + -sis and thrombo- + -sis.  --Lambiam 16:46, 21 October 2022 (UTC)

Interesting since θρόμβωσις is analyzed as θρόμβος + -σις, and this is given in OED as well. Thus, thrombosis would more analogically be surface-analyzed as thrombo- + -sis.

OED has -sis entry, covering e.g analysis and sepsis. MW has -sis covering peristalsis. sepsis is σήπω + -σῐς, so the surface analysis involving -sis seems fine.

We do have the option of reanalyzing -osis as -o- + -sis. It would be a bold option, deviating from sources. We will have to deviate from sources one way or another in at least a limited number of cases since their treatment of -oX is inconsistent. Some cases of this reanalysis should be less controversial: cytosis is now analyzed as cyto- + -osis, but would more naturally be cyto- + -sis. There is still the weird option of cyt- + -osis. gastrosis is the same case. --Dan Polansky (talk) 18:41, 21 October 2022 (UTC)

You write, “Thus, thrombosis would more analogically be surface-analyzed as thrombo- + -sis." By talking about reanalysis of terms like cyano- + -sis and thrombo- + -sis, I implied that this was the correct etymological analysis (modulo transcription) to start with; the suffix -osis emerged from an etymologically incorrect reanalysis.  --Lambiam 10:39, 22 October 2022 (UTC)

Keep and mark as non-productive or "mostly non-productive" as far as applicable. Another non-productive affix is -ion, and it serves a useful documentation function. --Dan Polansky (talk) 10:17, 21 October 2022 (UTC)

• RFD-kept: no consensus for deletion (WT:VPRFD). --Dan Polansky (talk) 10:52, 4 January 2023 (UTC)