Sum: `workaholic'

[Larry Trask]

----------------------------Original message----------------------------
The other day I posted a query requesting a name for formations like
`workaholic', `Camillagate' and `kissogram'.  Textbooks often class
these as blends, but they aren't, really: `workaholic' can't be
reasonably analyzed as `work' plus `alcoholic'.  Rather, a
morphologically arbitrary piece is ripped out of an existing word like
`alcoholic', `Watergate' and `telegram' and pressed into service as a
kind of affix.
 
One or two respondents suggested that this was just a form of
reanalysis.  Well, it is, sort of.  But, in the most familiar cases of
reanalysis, such as `hamburger' --> `cheeseburger' and `bikini' -->
`monokini', there seems to be a perception that the remaining element
(`ham-' or `bi-') is also a recognizable morpheme, which is not the
case here.  But maybe this is hairsplitting on my part.
 
Another respondent argued for `metanalysis'.  This term is only
familiar to me as a label for boundary shifts, as in `an ewt' --> `a
newt' and `a napron' --> `an apron'.  But the suggestion is that we
might generalize the term to cover false separations in general.
 
Yet another suggestion was `stump compound'.  But this term, in my
experience, denotes a formation obtained by combining arbitrary
initial stretches from the words in a phrase of identical meaning, as
in `sitcom', `sci-fi', German `Gestapo' and Russian `Sovnarkom'.  And
that's not what we have here.
 
There were other suggestions, some of them in print:
 
`constellation'
`local generalization'
`morphologization'
`remorphologization'
`analogical form'
`false separation'
`pseudo-suffixation'
`hybrid formation'
`creative compounding'
`meiosis'
 
This last one, a biological metaphor, looks interesting but apparently
hasn't appeared in print.
 
However, there have been some further terms used in print for exactly
the kind of formation I'm interested in.
 
First, Otto Jespersen coined `secretion' in his 1922 book _Language_.
I suppose this term deserves some kind of priority in English, but it
hasn't been used much, and it's not ideal: the imagery, whatever it
is, is opaque.
 
Second, the term `abstracted form' has appeared in print to label
these things, though I don't have the full reference.  I'm guessing,
though, that this label applies only to the detached piece, like
<-(o)holic>, rather than to a full word constructed with it.
 
Third, the term `Kontaminativkompositum' has been used in German.
This translates as `contaminative compound'.  But, apart from its
length, this term has problems: things like `workaholic' are not
strictly compounds, since they don't consist of free morphemes, and
moreover I don't find it easy to see where the contamination resides.
 
Fourth, the term `telescopage' has been used, presumably in French.
This translates as `telescoping', but again the imagery seems wrong.
As one respondent pointed out, `telescoping' would appear to be more
appropriate for formations like `glitterati'.  Maybe Jim Matisoff's
term `morphanization' is another example.
 
Fifth, Valerie Adams (and others?) have used the term `splinter' for
the extracted and re-used piece (like <-(o)holic>).  This suggests
`splinter formation' for the final result or for the process of
forming it, but apparently nobody has ever used this term.
 
One respondent pointed out that English allows so many individual
varieties of eccentric word-formation that it is very difficult to put
them into neat pigeonholes.  I'm afraid this is just true, and we'll
probably never have a perfect terminology for all cases, but I would
like to have an agreed term for this increasingly frequent type.
 
Several people drew attention to the importance of prosodic features
in these formations.
 
Finally, many respondents passed on some very interesting examples
from English, French, German, Russian, Croatian and Chinese (at
least).
 
My thanks to Joyce Tang Boyland, Margaret Winters, Tony Breed, Benji
Wald, Richard Coates, Michael Cysouw, Guido Mensching, Chris Hogan,
Gabriella Rundblad, Roger Lass, Chris Jeffery, Jim Rader, and Alemko
Gluhak.
 
Larry Trask
COGS
University of Sussex
Brighton BN1 9QH
UK
 
larryt at cogs.susx.ac.uk