Suppletion (summary)

[gladney at ILLINOIS.EDU]

Dear Suppletion Discussants,

Our string seems to have petered out, so let me offer a few summarizing remarks.

There seems to be broad agreement that suppletion his inversely related to phonology. Elena Arkhipova is quite clear on this point.  If _sosed_ and _sosed'_ occurred as word forms,  she writes, they would be as distinct as _mat_ and _mat'_.  So for her  'neighbor' (also 'devil') are suppletive nouns, the same as _rebenok_ and _deti_.  Gunter Schaarschmidt concurs: stem-final palatalization conditioned by the plural feature (_sosedu_ ~  _sosedjam_) is not regular, ergo two different stems, ergo suppletion.  Elena and Gunter's reasoning seems to be that morphemes consist of phonemes, so if /d/ and /d'/ are different phonemes /sosed/ and /sosed'/ have to be different morphemes.  Another possibility (my reference is Charles Hockett in vol. 37 of Language) is to say that morphemes don't consist of phonemes but of somethng called morphophonemes, {d, d'} being an example and {k, cz} as in _oko_ ~ _oczy_  another.  My objection to morphohonemes is that, unlike phonemes, they can't b!
 e represented as bundles of phonetic features.

With regard to _oko_ ~ _oczy_, Wayles Browne asks, "How much do the singular and plural stems have to be different before we call it suppletion?  Ewa W. responds that it is "no suppletion sensu stricto, because it is a regular change" (sort of, she adds later; there's also _oka_).  Natalia Kondrashova agrees: The change k > ch is a regular palatalization occurring throughout Slavic, and since it is caused by affixation (historically), it doesn't amount to suppletion. Dick DeArmond adds that suppletion was never intended to involve stem variants (at least those that appear to be phonologically related); if it did, Slavic would have too darn much of it.

Elena brings up the notion of paradigm, hence of inflection.  We may assume that Slavic speakers have root words like /vrem/, /ok/, /ljud/, and /beg/ and that they also have rules for inflecting them.  The rules for /vrem/ involve an /en/ element, those for /ljud/ involve an /yn/ in the singular but not in the plural, those for /ok/ and /beg/ involve not entirely predictable palatalizations.  But suppletion involves roots, and if it's the same root, then no suppletion.  

Regarding _vremja_ Elena speaks of a sub-stem.  But a sub-stem would differ from a stem, and having two amounts to suppletion, which I don't think she considers _vremja_ to involve.  She should just call it inflection and let it go at that.  By the way, Russian speakers who say _Skol'ko vremja?_ must be accepting Jakobson's analysis of the nom. sg. form as /vr'em'+o/, substituting the gen. sg. ending /a/, as in _slovo_ ~ _slova_.

In the same vein, Preslav Nakov cites Bulgarian _krilo_ 'wing', which when it refers to birds has an irregular nom. plural ending derived from the dual but for mechanical things that fly has the productive ending. This comes up also with Ru. _koleno_.  The Orthoepic dictionary solves the problem by saying there are several _koleno_ nouns, which differ in their plural forms.  This is like saying _fit_ ~ past tense _fit_ is a different verb than _fit_ ~ past tense _fitted_.  Or do /kril/ and /kolen/ have suppletive plural endings, like _ox_ and _box_?

Some would balk at putting the presence/absence of /yn/ with _ljudyna_ and of /in/ with _grazhdanin_, _anglichanin_, etc. in the same class as /k/ > /ch/ and /d/ > /d'/.  They are suffixes and have meaning (in the case of Ru. /in/, more in some occurrences than in others).  If suffixes are morphemes the same as root morphemes, their present/absence in a word form is as much a matter of suppletion as the presence/abseznce of roots.  But if this argument doesn't suit you, you can go with Curt Woolhiser's compromise: there's "weak suppletion" that involves suffixes and "strong" or full suppletion that involves roots.

Paradigms.  Paul Gallagher is right in defining suppletion as "a process whereby unrelated words
come to be used in a complementary arrangement", but i wish he hadn't added "... to form a single paradigm".  I think we need to distinguish between the gap in a paradigm on the one hand, e.g., no singular forms for _deti_ (by the way. thanks, Siniša Habijanec, for calling our attention to how speakers of Lower Sorbian deal with this gap), and how a  gap is coped with on the other.  I think it is arbitrary to promote to paradigm status some ways of coping and not others.  Where do you draw the line? Is _budu pobeditelem_ the 1st sg. of _pobedit_?  Is _to be able_ the infinitive form of _can_?  Natalia defines suppletion as the "substitution of the whole  ROOT with a different one", but I think she's putting undue emphasis on the coping part.  And I wish she wouldn't call it "a morphological change". That makes it sound like it's something done by a grammar component called Morphology.  However, Webster's New World Dictionary (3rd ed., 1988) gives as its third definition of !
 _suppletion_ "the morphological process by which such replacement [i.e., go+PAST > went] occurs".

So, Olga, you "would love to have a clarification of what suppletion really means". Looking the term up in the dictionary doesn't always help.  While it appears that most of us more or less agree it involves roots, not suffixes, in several dictionaries you will find as an example of suppletion the _-en_ in _oxen_ as opposed to the _-es_ of _boxes_.  By the way, how could you write "not being a linguist myself but merely a philologist"?  "Russkij filolog" is what Roman Jakobson had inscribed on his tombstone.

Please excuse the length  Thank you all for your patience.

Frank

-------------------------------------------------------------------------
 Use your web browser to search the archives, control your subscription
  options, and more.  Visit and bookmark the SEELANGS Web Interface at:
                    http://seelangs.home.comcast.net/
-------------------------------------------------------------------------