bound roots

[Laura Gonnerman]

Hi everyone,

To follow up on the comments of Marc Joanisse & others, I wanted to add that
in a variety of studies that I've carried out with Elaine Andersen, Mark
Seidenberg, and Dave Plaut, we've been able to show that speakers are in
fact sensitive to similarities in meaning and sound patterns between
morphologically related words. In a series of experiments, participants
showed priming effects that were modulated by the degree of overlap between
word pairs, with boldly-bold priming more than lately-late, which primes
more than the semantically dissimilar hardly-hard.
 
These effects hold for different types of words, e.g., two suffixed words
(saintly-sainthood) or prefixed-stem pairs (preheat-heat) or even words with
no 'morphemes' per se (snarl-sneer).

They also hold when the prime words are masked such that participants are
unaware of the primes.

And in recent imaging work with Joe Devlin, we showed that brain regions
that were active for morphologically complex items overlapped with those
active for semantically or orthographically related items. We found no
evidence for an area dedicated to 'morphology'.

 All that to suggest that speakers are influenced in processing by the
degree of semantic and phonological similarity between words, regardless of
whether they are aware of the morphological relationships between words.

As for the Œmit¹ words, there would certainly be behavioral effects for
permit-
permission type pairs, but likely not for most permit-submit pairs, which
may contain a
morpheme linguists recognize, but are too different in their meanings and
sound patterns to influence normal lexical processing.

For references beyond the one Marc mentioned, see also:

Devlin, J.T., Jamison, H.L., Matthews, P.M., & Gonnerman, L.M. (2004).
Morphology and the internal structure of words. Proceedings of the
National Academy of Sciences, 101 (41), 14984-14988.

Plaut, D.C. & Gonnerman, L.M. 2000.  Are non-semantic morphological effects
incompatible with a distributed connectionist approach to lexical
processing?  Language and Cognitive Processes, 15, 445-485.

Best,

Laura Gonnerman

*********************************
Laura M. Gonnerman
Assistant Professor
Department of Psychology

Chandler-Ullmann Hall Rm 339
17 Memorial Drive East
Lehigh University
Bethlehem, PA 18015

lag5 at lehigh.edu
phone: 610 758-4967
fax: 610 758-6277
*********************************

> Dear all -
>  
> I must admit that I am a bit puzzled by the discussion.  We have always taken
> it for granted, in linguistics, that knowledge of language does not refer to
> conscious knowledge of language.  Whether or not speakers think that a word is
> complex is no doubt an interesting question from many respects, but typically,
> we do not take any such reflections to be indicative of structure (for
> instance, we do not necessarily take seriously native speakers thoughts about
> whether sentences are complex or not or on why they may ill-formed).  What,
> ultimately, reflects knowledge, in the sense linguists typically mean it, is
> whether speakers linguistic behavior indicates the existence of such
> knowledge, at times not conscious.  For instance, if a native speaker of
> English has the intuition that the nominalized forms of commit, permit, emit
> etc. all involve the same allomorphic change (i.e., comiSSion, permiSSion,
> emiSSion), then it is at least suggestive of the fact that they know that
> 'mit' has properties which cut across all its occurrences, regardless of the
> fact that 'mit' is not a word and that statistically, most speakers may not
> judge a word as 'permit' to be complex.  Similarly, of course, for 'ceive', as
> in 'reception', 'conception', 'inception', 'perception' etc.
>  
> Hagit Borer
>  
>  
>>  
>> ----- Original Message -----
>>  
>> From:  Benjamin Munson <mailto:munso005 at umn.edu>
>>  
>> To: Carolyn Chaney <mailto:cchaney at sfsu.edu>  ; info-childes
>> <mailto:info-childes at mail.talkbank.org>
>>  
>> Sent: Friday, November 12, 2004 7:39  AM
>>  
>> Subject: Re: bound roots
>>  
>> 
>> Dear List-Mates:
>> 
>> I feel compelled to chime in.  I  would like to echo Marc Joanisse's
>> statement that the difference between  decomposed and non-decomposed words is
>> continuous rather than discrete.   In addition to the reference he suggested,
>> I would point you to the research  of Jen Hay.  She showed that the extent to
>> which a derived word can be  decomposed into a root-plus-derivational
>> morpheme is dependent on the relative  frequency of the stem and the derived
>> form.  This is illustrated in the  following article:
>> 
>> Hay, Jennifer (2001) Lexical Frequency in  Morphology: Is Everything
>> Relative? Linguistics , 39 (6), 2001, pg  1041-1070.
>> 
>> I would also point you to other relevant papers by  her:
>> 
>> Hay, Jennifer (2002) From Speech Perception to Morphology:  Affix-ordering
>> Revisited.  Language 78.3, 2002: 527-555.
>> Hay,  Jennifer and Ingo Plag (2004)  What constrains possible suffix
>> combinations?  On the interaction of grammatical and processing  restrictions
>> in derivational morphology.   Natural Language and  Linguistic Theory 22:
>> 565-596 
>> 
>> I'm not sure that detailed analyses  in these articles clearly inform
>> language pedagogy (re Caroline's original  question), but they are nifty and
>> relevant works nonetheless, and they serve  to bolster Marc's point about
>> these differences being continuous.
>> 
>> Cordially,
>> Ben Munson
>> Asst. Prof., Dept. of  Speech-Language-Hearing Sciences
>> University of Minnesota,  Minneapolis
>> 
>> At 05:49 PM 11/11/04, Carolyn Chaney wrote:
>>  
>>> In my Language for Teachers class  we were discussing various kinds of
>>> morphemes, and we discovered that we  had difficulty knowing if certain
>>> words were free morphemes or a combo of  an affix plus a bound root.  This
>>> was particularly difficult when  the word has a syllable that looks like am
>>> affix, such as mothER or  DEcide.  Cases where there are several like words
>>> (receive, deceive,  conceive) look like bound roots.  Mother seems clearly
>>> to be a free  morpheme, as a mother is not one who moths.  But what  about
>>> decide?  inept?  nonchalant?  uncouth?   refine?  Uncouth, for example, is
>>> given in texts as an affix plus  bound root, but surely it doesn't mean
>>> not-couth.  Does anyone have  a clear explanation of how to distinguish
>>> words with affix-looking parts  from words that really have affixes + bound
>>> roots, preferably an  explanation that does not require looking up
>>> derivations in the  dictionary?
>>> 
>>> Thanks for the help!
>>> 
>>> Carolyn Chaney
>>> Just call  me stumped
>