Content uploaded by Paul Edmunds
Author content
All content in this area was uploaded by Paul Edmunds on Nov 30, 2021
Content may be subject to copyright.
Content uploaded by Paul Edmunds
Author content
All content in this area was uploaded by Paul Edmunds on Nov 30, 2021
Content may be subject to copyright.
Buen- -- Buena Gente:
Repair in the Spanish of the Southwest
Paul Edmunds
University of New Mexico
1. Introduction
Repair is a phenomenon which we witness every day, consciously or unconsciously, as we interact
in conversations. Repair has been defined as the phenomenon in which an emerging utterance is
truncated and then restarted or continued with some alteration in the syntactic form, if only including
repetition of the repaired segment (Fox & Jasperson 1995, Sanchez-Ayala 2003). There are different
types of repair. In replacing repair, an emerging word is truncated and a different word is used in its
place. In recycling repair with modification, some parts of the repaired segment are repeated but other
parts are replaced or modified (Fox & Jasperson 1995, Schegloff 1979). Another common type of
repair is pure recycling (Sanchez-Ayala 2003). In pure recycling, the repaired segment (the word that
is truncated and restarted) and the repairing segment involve the same lexical item(s). Examples 1 and
2 below show two instances of pure recycling, the focus of the present paper.
(1) Oh es -- es-- Colorado es otro estado. [10-8A, 119-120]
'Oh it’s – it’s – Colorado is another state.'
(2) No entendía inglés pero en- -- entendí lo que ella me dijo y todo. [10-8A, 354-355]
'I didn’t understand English but I un- -- I understood what she told me and everything.'
Example 1 shows a case of pure recycling appearing between two completed words, a
phenomenon sometimes called a repeat (Maclay & Osgood 1959). In this case, an entire lexical item is
uttered and then repeated completely. Example 2 is an instance of recycling occurring within the word
in progress, often referred to as a false start (Altmann 1997, Maclay & Osgood 1959). In this case, the
word in progress is truncated at some point and subsequently reformulated completely immediately
after the truncation. In both cases, the particular stretch of discourse in question has often been
considered a disfluency which may or may not inhibit the comprehension of the listener (Altmann,
1997, Clark & Wasow 1998, Fox Tree 1995, Maclay & Osgood 1959). The idea of repair as a
‘disfluency’ is not, of course, held by everyone, as we will see further on.
The present study will consider instances of pure recycling at the single word level to determine
which parts of speech are most commonly involved in repair. This study hopes to make an important
contribution to the field of linguistics by showing how patterns of repair occur at a more micro- (word)
level and in a language which has received little attention regarding this phenomenon, Spanish.
Research has been done on repair examining the syntactic level of the repaired segment. A study
of repair in Madrileño Spanish (Sanchez-Ayala 2003) found that recycled segments most frequently
appear locally. That is, repair occurs within phrasal constituents rather than repairing back to the
beginning of the main clause in progress. In addition, Sanchez-Ayala found that prepositional phrases
most frequently undergo repair, followed by verbal complexes, postverbal NP arguments and to a
much lesser extent adjective and adverbial phrases.
* Thanks to Catherine Travis who provided valuable direction on a previous version of this paper, and the two
anonymous reviewers for their helpful comments.
© 2006 Paul Edmunds. Selected Proceedings of the 8th Hispanic Linguistics Symposium, ed. Timothy L. Face
and Carol A. Klee, 204-213. Somerville, MA: Cascadilla Proceedings Project.
Fox and Jasperson (1995) also found that repair is not chaotic but rather has syntactic
organization. These researchers argue that, as syntax is created during the course of conversation,
syntax must allow for the possibility of seemingly ‘ungrammatical’ segments and thus, repair. In
addition, Fox and Jasperson (1995) found that repair in copular verb constructions behaves differently
than in regular verb phrases in that repair involving copulas seems to have no significant constraints
while regular verb phrases normally experience repair back to the beginning of the finite clause.
In a cross-linguistic study comparing repair at the syntactic level in English and Japanese, Fox et
al. (1995) note that the language in question organizes repair according to its own syntactic constraints
and furthermore the speakers of the languages use repair interactionally for specific purposes. Thus,
different languages may show different patterns of repair but should not show random treatment of the
phenomenon. Again, repair is not seen as a disfluency but rather is used as a strategy to achieve
interactional goals.
Finally, a study by Maclay and Osgood (1959), which looked at the kinds of words involved in
repair, found that false starts (i.e., repair occurring within a word) generally involve lexical items (i.e.,
nouns and verbs) whereas repetitions of a single word, or “repeats” (i.e., repair occurring between
words) is generally found in function words (i.e., prepositions and articles). Maclay and Osgood
(1959) found that 71% of the repair in their study involved a single word, while only 17% involved
more than one word and the remainder involved only a single phoneme or syllable. In addition, these
researchers found that repair generally occurs between constituents rather than within one constituent.
The present study analyzes repair in the form of pure recycling in the Spanish of the southwestern
United States, an area where Spanish/English bilingualism is prevalent. As such, this study, although
centered on repair in Spanish, can add some cross-linguistic evidence to the literature on repair by
analyzing instances in which repair precedes the use of an English-origin word, either through code-
switching or borrowing. Silva-Corvalán (2001:272) has noted that use of an English word while
speaking in Spanish can be the result of the necessity of filling a lexical gap. She argues that the use of
an item in the non-matrix language may be utilized because the word being searched for in the matrix
language is infrequent and therefore difficult to access. To lower the cognitive processing load, then,
the speaker may use a word from the non-matrix language (Silva-Corvalán 2001:290). Silva-Corvalán
(2001:272) also notes that the use of the non-matrix item may be realized because that item is more
specific, precise or more transparent than any available word in the matrix language. The present study
will consider these factors when looking at instances of the use of English words incorporated into the
matrix language, Spanish, and examine if a higher instance of repair occurs in these situations.
As noted earlier, much of the previous research on repair has been done at a larger, syntactic level.
The present study is based primarily at the word level in order to investigate the types of words which
are involved in pure recycling, thus providing an opportunity to study this phenomenon at a more
micro level. Repair at the single word level was adopted for this study as an initial count of repair in
the southwestern data (New Mexico-Colorado Spanish Survey, Bills & Vigil 1999) revealed that
approximately 75% of all repair involved a single word, roughly the same percentage that Maclay and
Osgood (1959) found in their study with English data. In particular, the study seeks to determine if
function words or lexical items are most commonly involved in repair in southwestern Spanish and if
these words conform to a closed class list or an open class list. In addition, the study will examine if
the repair occurs more frequently between two words or within a word and how often repair precedes
the use of the non-matrix language. Through analyses of these variables, the study will also seek to
determine if a pattern for repair at the word level exists and, if so, will try to answer the question of
what motivates this pattern.
2. Methodology
Interviewee responses from six interviews from the New Mexico-Colorado Spanish Survey
(NMCOSS, Bills & Vigil 1999) were analyzed for occurrences of repair in the form of same-turn,
single-word pure recycling. The 16 data files for these six interviews totaled approximately 21,000
words, excluding interviewer speech. The interview format of the data produced guided yet informal
conversations between the interviewer and interviewee. The interviewees in these conversations are all
native speakers of Spanish who, in many cases, also demonstrate a high level of English proficiency.
205
The interviewees represent three men and three women ranging from 45 to 88 years of age. The
speakers received between two and sixteen years of education and worked mainly in the home or in
working-class occupations. The interviews of the NMCOSS took place in a wide variety of areas
across New Mexico and Colorado, however only New Mexican data is represented here.
A total of 271 instances of pure recycling repair from the interviewees were coded in Excel and
analyzed. These tokens were found by loading the corpus files into Monoconc (a concordance program
useful for drawing frequency distributions) and searching for truncation markers in the corpus data. In
addition to coding for part of speech (POS), the number of syllables in the repairing word were coded
to determine the typical length of words involved in repair and to investigate the question of why a
word would be more or less likely to experience constituent-internal repair. Furthermore, the
constituent space of the repair was coded as either within the word or between words. For instance (3)
was coded as constituent-internal whereas (4) was coded as occurring between constituents. Finally,
instances of use of an English-origin word occurring immediately after or nearly immediately after an
instance of repair were also coded to analyze how often instances of pure recycling precede the use of
the non-matrix language.
(3) Hac- -- hacían mucha ropa.
'They ma- -- they made a lot of clothes.' [144-3A2, 008-009]
(4) Yo nunca – nunca tomé el curso.
'I never took the course.' [102-1A1, 222-223]
To be able to understand the percentages of repair by POS, it was necessary to determine how
many words by POS were actually present in the data. To do so, a sample of the interviewee only data
from the 16 corpus files was created by loading all 16 data files into Monoconc and obtaining the
frequency distribution of all words in the corpus. The results of the word frequency distribution were
arranged in order of most frequently occurring to least frequently occurring. These results were then
saved as a text file and imported into Excel. In Excel, the sample was limited to words that appeared in
the corpus at least 5 times. Of the 21,278 total words in the corpus, the number of individual words
with a frequency of appearance of at least five, including verb conjugations, was 375. These 375 words
accounted for 11,591 total words in the corpus, or 54.5% of the corpus data. These 375 words were
then coded in Excel for POS and later used to compare the frequency of instances of repaired words to
non-repaired words by POS (See Table 1). Words excluded from coding were those which could not
be ascertained for POS by examining the words independent of their context. Examples of excluded
words are chiquitas (which could be an adjective or a plural noun) and trabajo (which could be a noun
or verb). Furthermore, discourse markers such as ahm, mhm, and eh were not included in the study.
3. Results
Table 1 shows the distribution of the repair by part of speech. To be able to compare the number
of words repaired by POS, the number of words not involved in repair is included, as is the combined
total of repaired and non-repaired tokens. The percentage of repair by POS, arranged from most to
least frequently occurring, is listed in the far right column.
206
Part of Speech Repaired N
Repaired N not
repaired Total %
repaired
Preposition 61 1493 1554 3.9
que (relativizer, question, subjunctive) 19 508 527 3.6
Article 38 1160 1198 3.2
Adverb 63 1986 2049 3.1
Possessive Pronoun 6 202 208 2.9
Conjunction 35 1475 1510 2.3
Subject Pronoun 8 492 500 1.6
Copula 3 268 271 1.1
Adjective 8 719 727 1.1
Verb 19 1743 1762 1.1
(Reflexive) Clitic Pronoun 5 699 704 0.7
Noun 6 1095 1101 0.5
Total 271 11840 12111 2.2
Table 1: Distribution of repair by POS repaired1
Table 2 shows the parts of speech repaired and their appearance either word-initially or between
words. This table also includes the average number of syllables per word for each POS. Table 2 is
organized by the frequency of occurrences of repair between words, from most frequently to least
frequently occurring.
Part of Speech Repaired N
Repaired # Between %
Between # Within % Within Avera
g
e #
syllables
Conjunction 35 35 100.0 0 0.0 1.1
que (rel., question, subj) 19 19 100.0 0 0.0 1
Adverb 63 58 92.1 5 7.9 1.4
Preposition 61 55 90.2 6 9.8 1.1
Article 38 34 89.5 4 10.5 1.1
Subject Pronoun 8 7 87.5 1 12.5 1.6
Possessive Pronoun 6 5 83.3 1 16.7 1
(Reflexive) Clitic Pronoun 5 4 80.0 1 20.0 1
Copula 3 2 66.7 1 33.3 1
Verb 19 8 42.1 11 57.9 3
Adjective 8 0 0.0 8 100.0 1.9
Noun 6 0 0.0 6 100.0 2.8
Total 271 227 44
1.5
Table 2: Distribution of repair by place in word
Table 3, included in section 4.3, shows instances of repair preceding the use of an English-origin
word, by way of code-switching or borrowing.
1 Only one case of repair involving a past participle was found (hecha). That entry was included with the category
of adjectives. Similarly, the one occurrence of repair in a demonstrative subject pronoun (ése) was included with
the category of subject pronouns.
207
4. Discussion
4.1. Analysis by POS repaired
One striking result seen in Table 1 is the small amount of single-word pure recycling present in
the data. Only a total of 2.2% of all the interviewee corpus data is involved in single-word repair.
Table 1 also shows that, in general, the amount of repair can be divided by categories of content and
function words and that only the function words (prepositions, que, articles, adverbs,2 possessive
pronouns, and conjunctions) represent over 2% of repair by POS. In the following section a further
analysis of function and content word items involved in repair will be treated.
4.1.1. Repair in function words (closed-class items)
Table 1 shows that the parts of speech with the highest frequency of repair are all function words:
prepositions (3.9%), que (3.6%), articles (3.2%), adverbs (3.1%), possessive pronouns (2.9%),
conjunctions (2.3%), and so on. Function words are more grammaticized in that they are more
frequent, generalized and abstract. They also are smaller constituents which project larger constituents
like nouns and verbs. The six aforementioned function words alone represent 222 of the 271 instances
of repair in the study, or 82%. Below examples of repair attested in some of the more frequently
repaired function words are given, beginning with the most frequently occurring items.
Prepositions (N=61)
(5) Por ejemplo en -- en el valle era el.. Día de San Miguel. [144-3B2, 496-497]
'For example in – in the valley it was St. Michael’s Day.'
(6) Pero yo estaba muy contenta haciéndolo por -- por mis hermanos. [117-1A3, 378-379]
'But I was very happy doing it for – for my brothers.'
Prepositions accounted for the greatest amount of repair by POS (3.9%) and show a tendency for
repair between words (90.2%, see Table 2). This tendency for repair between words is most likely due
to their highly monosyllabic nature as seen by their low average number of syllables (1.4).
Articles (N=38)
(7) Ese me trujo también un -- un canasto. [10-8A, 534-535]
'That one also brought me a – a big basket.'
(8) Y ése -- ése sí está traducido en inglés y en español. [102-3A2, 238-239]
'A that one – that one is translated into English and into Spanish.'
Articles were frequently involved in repair (3.2%). Again their tendency for repair between words
(89.5%) was due to their nearly monosyllabic nature (average number of syllables = 1.1).
Adverbs (N=63)
(9) Yo nunca -- nunca tomé el= .. curso. [102-1A1, 222-223]
'I never – never took the= .. course.'
(10) Nunca hemos tenido una tradición como -- como= muchos países tienen tradiciones. [144-
4B2, 122-123]
'We’ve never had a tradition like – like= a lot of countries have traditions.'
2 In the present data, all of the tokens of adverbs found to participate in repair are function words (i.e., cuando,
donde, no, ya)
208
Adverbs accounted for 3.1% of single-word repair. In total, 92.1% of adverbs showed repair
between words, although their average number of syllables per word was only 1.4.
Conjunctions (N=42)
(11) Vinieron entre los primeros pobladores porque -- porque los perseguían. [144-4B2, 435-436]
'They came among the first settlers because – because they were being persecuted.'
(12) Pasteles y= cositas así y -- y también algunas veces un poco más. [144-3B2, 294-295]
'Cakes and= little things like that and – and also sometimes a little more.'
Conjunctions accounted for 2.3% and showed 100% of all instances of repair between words, due
again to their highly monosyllabic nature (average number of syllables = 1.1).
To summarize, function words accounted for 238 of the 271 cases of repair in the present data, or
87.8%. Of these 238 function words involved in repair, 92% occurred between words. The average
number of syllables per word for function words was 1.1.
4.1.2. Repair in content words (open-class items)
Of all of the tokens of repair in the study, only 12.2% are content words. Of these 33 tokens, 25
experience word-internal repair (75.8%). The average syllable length of the content words involved in
repair is 2.6. Below are examples of repair for the open-class items found in the NMCOSS data: verbs,
adjectives and nouns. Repair in copular constructions with the verb ser is also included in this section.
Verbs (N=19)
(13) Hac- -- hacían mucha ropa. [144-3A2, 008-009]
'They ma- -- they made a lot of clothes.'
(14) En- -- Entiendo más de lo que hablo. [20-1A1, 324-325]
'I un- -- I understand more than I speak.'
As can be seen in Table 1, verbs were rarely involved in recycling, appearing only 1.1% of the
time compared to their non-repaired usage. Their higher syllable count (average = 3, see Table 2) lent
this word type to a higher amount of word-internal repair (57.9%).
Copula ser (N=3)
(15) Pero el Antonio me acuerdo que era -- era medio abogadito tambíen. [144-4A2, 183-184]
'But I remember that Antonio he was – he was sort of a little lawyer too.'
(16) Oh es -- es-- Colorado es otro estado. [10-8A, 119-120]
'Oh it is – it is – Colorado is another state.'
(17) Nuestra fe e- e- es una cosa im- -- [144-4B2, 095]
'Our faith i- i- is something in- --'
Two of the three copular constructions with ser showed repair between words (examples 15 and
16) while one instance of repair occurred word-internally (example 17). Again, the monosyllabic
nature of these tokens favors between-word repair, although word-internal repair is possible and does
occur in monosyllabic words. Due to the small amount of tokens of repaired copular constructions, it is
not possible to comment on the findings by Fox and Jasperson (1995) that copular repair does not tend
to follow any certain patterns. However it is interesting to note that only three instances of repair
209
involving the copula ser were encountered, far fewer than the number of main verbs which were
involved in repair.
Adjectives (N=8)
(18) Oh= los p- -- primero por supuesto. [144-4B2, 148-149]
'Oh= the f- -- first ones of course.'
(19) Pues ahí estaba .. li- -- libre. [144-3B2, 146-147]
'Well there he was .. fr- -- free.'
As with the verbs, adjectives also showed only 1.1% of involvement in repair compared to the
number of adjectives which did not undergo repair. Interestingly, adjectives showed 100% occurrence
of repair within words even though their average number of syllables was quite low (1.9). Only three
of the eight adjectives involved in repair were monosyllabic (dos, tres).
Nouns (N=6)
(20) Eran todos ho- -- hombres. [20-1A1, 238-239]
'They were all me- -- men.'
(21) Y cuando mis he- -- hermanos vinían en la escuela, ellos comían. [117-1A3, 231-232]
'And when my bro- -- brothers came from school, they would eat.'
As we saw above in the adjectives, nouns too showed 100% of repair word-internally, although
their average syllable count was more similar to that of verbs (2.8 syllables/word).
4.2. Summary of POS and place of repair
Function words show the highest frequency of repair and the most instances of repair between
words. Of the function words, 219 of 238 experience repair between words, or 92%. This figure,
however, is not surprising when considering that the average syllable length of these function words is
1.1. The monosyllabic nature of the words obviously does not lend itself well to creating repair within
a word. Although repair word-internally was found to be possible, it was quite rare. Furthermore, the
low semantic value of function words may allow the speaker a more convenient opportunity to stall for
time and plan the following segment of speech before arriving to the semantically heavier words. This
may suggest that function words, as smaller constituents, are used as an ideal place for a speaker to
select or plan the following, larger constituents such as nouns and verbs. Further investigation on this
topic could address this possibility from a cognitive standpoint more directly.
When examining the content words, we see that 24.2% of repair appears between words. Of the 25
verbs, adjectives and nouns involved in repair, 75.8% of the instances of repair occur word-internally.
For verbs, we see that just over half of all repair was found within the word in progress (11/19, or
58%) and for adjectives and nouns, every occurrence of repair was found word-internally (8/8 and 6/6,
respectively). I propose that repair is found more word-internally in the case of nouns, verbs and
adjectives because they are longer when considering the syllable length. The average syllable length of
the nouns, verbs and adjectives together was 2.6. Clearly, there is more phonetic and morphological
substance to these words and therefore they have a higher possibility to experience repair word-
internally.
210
4.3 Repair preceding use of an English-origin word
Considering the strongly bilingual population of the southwestern United States, and in large part
of the speakers represented in this study, it is important to consider how repair interacts with bilingual
linguistic phenomena such as code-switching and borrowing. That is, does repair occur frequently in
situations in which the speaker works between both Spanish and English? Thus, the frequency of
single-word pure-recycling preceding the use of an English-origin word, either in the form of code-
switching or borrowing, was studied (Table 3).
Part of Speech Repaired English-Origin Word
Preposition 3
Article 3
Adverb 1
Conjunction 1
Verb 1
Possessive Pronoun 0
Subject Pronoun 0
Copula 0
Adjective 0
(Reflexive) Clitic Pronoun 0
Noun 0
Total 9
Table 3: Repair preceding an English-origin word
Clearly there are few instances in which repair preceded the use of an English-origin word (only 9
cases of the 271 cases of repair analyzed, or 3.3%). Prepositions and articles, once repaired, each
preceded 3 occurrences of an English-origin word, while adverbs, conjunctions and verbs preceded
only one instance each. The nine cases of repair involving a subsequent English-origin word or phrase
are seen below. Note that only two of the six speakers exhibited this phenomenon and, furthermore,
seven of the nine cases (77.8%) were uttered by the same speaker. The nine cases of repair preceding
an English-origin word are shown below in examples 22-30.
(22) Pero pronto pescas el – el lingo no? [102-1A1, 241-242]
'But soon you catch the – the lingo you know?'
(23) Yo gradé del -- del high school no? [102-2B1, 092-093]
'I graduated from – from high school, you know?'
(24) Porque teníamos munchos ah -- munchos ahm – neighbors. [117-1A2, 102-104]
'Because we had lots ah – lots ah – neighbors.'
(25) El -- el language de ellos era .. inglés. [117-1A2, 178-179]
'The – the language of theirs was .. English.'
(26) Tenían su class de—de—de—de bilingual. [117-1A2, 398-401]
'They had their - - - - bilingual class.'
(27) Ya el eh -- el eh -- Ronny and Leroy were already in high school. [117-1A3, 518-519]
'Already the eh – the eh – Ronny and Leroy were already in high school.'
211
(28) Que – que nunca I never gave up. [117-1A3, 616-617]
'That – that never I never gave up. '
(29) Porque tuve – tuve three girls. [117-1A3, 640-641]
'Because I had – I had three girls.'
(30) En – en – ah=, nineteen eighty five. [117-1A3, 663-665]
'In – in – ah=, nineteen eighty five.'
As noted earlier, Silva-Corvalán (2001) has proposed that use of an English-origin word may be
motivated by the existence of a lexical gap in the matrix language or for psycholinguistic factors such
as difficulty in remembering a low frequency word in language A and therefore using a word from
language B as a way to reduce the cognitive load involved in producing the utterance. Furthermore, a
speaker may use an English-origin word for ease of expression. Considering examples 22-30 above,
we see that the English words used, in general, have everyday counterparts in Spanish, with the
notable exceptions of examples 22 (lingo) and 28 (gave up) which have a jargon and phrasal-verb
quality, respectively. Given that the majority of the examples do not represent low frequency words in
Spanish, it is safe to say that the use of an English word in these cases was, in general, not done to fill
lexical gaps. Considering these findings, I would argue that single-word pure-recycling does not
predict the use of an English-origin word and that the instances shown in 22-30 above are most likely
done to ease the cognitive load of the expression.
5. Conclusion
Function words are involved in repair more than content words, likely due in general to their low
semantic load. In addition, repair involving function words occurs mainly between constituents, almost
certainly due to the short, mostly monosyllabic quality of these words. These findings are very similar
to those of Maclay and Osgood (1959) from their study of repair in English. The higher frequency of
repair in function words may indicate that a speaker is taking advantage of the low semantic quality of
items such as prepositions to initiate repair in an attempt to gain time to plan for the following segment
of speech of higher semantic value. This would help explain why content words experience much less
repair in the form of pure-recycling: simply put, once these words have begun to be uttered they have
already been premeditated sufficiently as to not require repair mid-utterance. Pauses required for
planning of subsequent speech can be mediated by repair in the function words. The fact that lexical
items show more word-internal repair, especially in the case of verbs, nouns and adjectives, is likely
due to the fact that they are longer words, on average more than twice the length in syllables as the
closed-class words investigated in this study. With the longer phonetic quality of the words often
comes a greater amount of morphology which lends itself to the possibility of word-internal repair.
Finally, these data show that cases of single-word pure recycling precede very few occurrences of
English-origin words. This result does not suggest that repair is done because of uncertainty of the
Spanish word. Due to the low number of tokens in this study, however, this result requires further
investigation.
In conclusion, the level of grammaticization, semantic load and length of words appear to predict
patterns of single-word repair in Spanish of the Southwest. Just as there is organization of repair at the
syntactic level as found in previous research, the results here indicate that repair is not random across
parts of speech and word classes but that there is a pattern to repair at the lexical level as well.
References
Altmann, Gerry T. M. 1997. The ascent of Babel. Oxford: Oxford University Press.
Bills, Garland D. and Neddy A. Vigil. 1999. Ashes to ashes: the historical basis for dialect variation in New
Mexican Spanish. Romance Philology 53.43-66.
212
Clark, Herbert H., and Thomas Wasow. 1998. Repeating words in spontaneous speech. Cognitive Psychology
37.201-42.
Fox, Barbara A.; Makoto Hayashi; and Robert Jasperson. 1995. Resources and repair: a cross-linguistic study of
syntax and repair. Interaction and grammar, ed. by Elinor Ochs, Emanuel A. Schegloff, and Sandra A.
Thompson, 185-237. Cambridge, UK: Cambridge University Press.
Fox, Barbara A., and Robert Jasperson. 1995. A syntactic exploration of repair in English conversation.
Alternative linguistics: descriptive and theoretical modes, ed. by Phillip W. Davis, 77-134. Amsterdam: John
Benjamins.
Fox Tree, Jean E. 1995. The effects of false starts and repetitions on the processing of subsequent words in
spontaneous speech. Journal of Memory and Language 34.709-38.
Maclay, Howard, and Charles E. Osgood. 1959. Hesitation phenomena in spontaneous English speech. Word
15.19-44.
Sanchez-Ayala, Ivo. 2003. Patterns of repair and constituency in Spanish conversation. Paper presented at the 6th
Hispanic Linguistics Symposium, Albuquerque, New Mexico.
Schegloff, Emanuel A. 1979. The relevance of repair to syntax-for-conversation. Syntax and Semantics 12.261-
86.
Silva-Corvalán, Carmen. 2001. Sociolingüística y pragmática del español. Washington DC: Georgetown
University Press.
213
Selected Proceedings of the
8th Hispanic Linguistics Symposium
edited by Timothy L. Face and Carol A. Klee
Cascadilla Proceedings Project Somerville, MA 2006
Copyright information
Selected Proceedings of the 8th Hispanic Linguistics Symposium
© 2006 Cascadilla Proceedings Project, Somerville, MA. All rights reserved
ISBN 1-57473-408-3 library binding
A copyright notice for each paper is located at the bottom of the first page of the paper.
Reprints for course packs can be authorized by Cascadilla Proceedings Project.
Ordering information
Orders for the library binding edition are handled by Cascadilla Press.
To place an order, go to www.lingref.com or contact:
Cascadilla Press, P.O. Box 440355, Somerville, MA 02144, USA
phone: 1-617-776-2370, fax: 1-617-776-2271, e-mail: sales@cascadilla.com
Web access and citation information
This entire proceedings can also be viewed on the web at www.lingref.com. Each paper has a unique document #
which can be added to citations to facilitate access. The document # should not replace the full citation.
This paper can be cited as:
Edmunds, Paul. 2006. Buen- -- Buena Gente: Repair in the Spanish of the Southwest. In Selected Proceedings of
the 8th Hispanic Linguistics Symposium, ed. Timothy L. Face and Carol A. Klee, 204-213. Somerville, MA:
Cascadilla Proceedings Project.
or:
Edmunds, Paul. 2006. Buen- -- Buena Gente: Repair in the Spanish of the Southwest. In Selected Proceedings of
the 8th Hispanic Linguistics Symposium, ed. Timothy L. Face and Carol A. Klee, 204-213. Somerville, MA:
Cascadilla Proceedings Project. www.lingref.com, document #1267.
