Intrasentential Codeswitching as Conceptual Projection of Lemmas
in the Bilingual Mental Lexicon
Longxing Wei
Montclair State University
This paper explores the bilingual speech containing intrasentential
codeswitching (ICS) in relation to the nature and activity of the bilingual mental
lexicon. ICS is defined as the use of two or more linguistic varieties within the
sentence boundaries in the same discourse. It explains the phenomenon of ICS
in terms of the abstract notions of ‘lemmas’ and ‘lemma activation.’ Lemmas
are abstract entries in the mental lexicon which contain semantic, pragmatic,
syntactic, morphological, and phonological information about actual lexemes.
However, lemmas in the bilingual mental lexicon are tagged for specific
languages. In this paper ICS is regarded as conceptual projection of languagespecific lemmas in the bilingual mental lexicon. What occurs in ICS leads to
hypotheses about principles structuring and governing the bilingual mental
lexicon and speech production process. The research findings based on ICS
involving language pairs such as Chinese/English and Japanese/English indicate
that ICS cannot be accounted for at the surface level of investigation. Rather,
any ICS phenomenon depends on bilingual cognitively based operations of an
abstract nature. The findings provide evidence that language-specific lemmas in
the bilingual mental lexicon are not equally activated during the process of ICS:
Some are conceptually activated, but others are not. It is only conceptually
activated lemmas of the Embedded Language (EL) (i.e., the guest language)
which can be switched into the sentential frame of the Matrix Language (ML)
(i.e., the base or host language). The naturally occurring ICS instances discussed
in this paper also provide evidence that there is differential activation of
participating languages in ICS in terms of language-specific lemma activation
for content morphemes, and language-specific lemma activation for
morphosyntactic procedures. However, activated EL lemmas must be
sufficiently congruent with ML counterparts at several levels for possible ICS
realizations. This paper proposes a bilingual lemma activation model of ICS.
Journal of Cognitive Science 6: 149 - 179, 2006.
ⓒ2006 Institute for Cognitive Science, Seoul National University.
150
Longxing Wei
Keywords: lemma, mental lexicon, conceptual, activation, morpheme,
congruence.
1. Introduction
Intrasentential codeswitching (ICS for short) is the alternative use of two
languages in the same sentence. Linguists examine switch-points: where the
switch is structurally possible, and switch items: what items from another
language can be switched. Unlike most previous studies of ICS which focused
on describing surface configurations of the so-called ‘mixed’ bilingual speech
without explaining the cognitive and linguistic motivations for such a bilingual
behavior, this paper explores the bilingual speech production process and
morphosyntactic constraints on ICS in relation to the nature and activity of the
bilingual mental lexicon. It specifically deals with the issues of unequal
activation of lemmas and lemma congruence checking procedures between the
language pairs involved in ICS. Based on some naturally-occurring ICS
instances from language pairs like Chinese/English, Japanese/English,
English/Chinese, and English/Japanese, this paper presents four arguments: (1)
Although, as generally conceived, there is only one mental lexicon for
bilinguals, lemmas contained in the bilingual mental lexicon are languagespecific. Although conceptual representations are shared, lexical
representations are independent across languages. (2) Lemma activation itself
is one of the levels of speech production. It receives input from the
Conceptualizer at the conceptual level and provides output for the speech
production Formulator at the functional level (cf. Levelt, 1989). It is in this
sense that we say that lemmas are conceptually projected from the bilingual
mental lexicon. (3) Although morphosyntactic procedures activated by lemmas
from the bilingual mental lexicon must involve parallel processing, the
bilingual’s languages involved in ICS are not equally activated at the same
time, and unequal activation of lemmas in the bilingual mental lexicon is one
of the major cognitive and linguistic motivations for ICS. (4) Bilinguals can
access any language known as the guest language in a discourse involving ICS,
but lemmas conceptually projected from that language must be congruent with
the base language counterparts regarding three levels of abstract lexical
structure: lexical-conceptual structure, predicate-argument structure, and
Intrasentential Codeswitching as Conceptual Projection of Lemmas in~
151
morphological realization patterns.
The major assumption underlying this paper is that ICS itself is a linguistic
system, and, like any linguistic system, it is governed and constrained by a set
of morphosyntactic principles and rules. The so-called mixed speech
phenomenon cannot be accounted for without exploring the nature of the
bilingual mental lexicon and the role of conceptual projection of lemmas in the
bilingual mental lexicon during the speech production process. The research
findings provide evidence that lemmas in the bilingual mental lexicon are not
equally activated: some are conceptually activated, but others are not. It is
conceptually projected lemmas together with lemma congruence checking
procedures between the language pairs involved in ICS which are the major
sources of structural constraints and cognitive motivations for such a bilingual
behavior. Based on this evidence, this paper proposes a model of bilingual
lemma activation in bilingual production. It concludes that conceptual
projection of lemmas in the bilingual mental lexicon results in the choice of
discourse modes and the selection of morphemes in ICS, and lemma
congruence checking becomes necessary for possible ICS configurations.
2. Conceptual projection of lemmas
Lemmas are abstract entries in the mental lexicon which support the surface
realization of actual lexemes. The lexical items stored in the mental lexicon not
only specify the meanings of words but also contain syntactic, pragmatic,
morphological, and phonological information about them (Levelt, 1989). Thus,
the mental lexicon is defined as the speaker’s internal representation of
language specific knowledge about the surface forms.
In speech production, the first step is to retrieve appropriate words from the
speaker’s mental lexicon. For each item, the mental lexicon contains its lemma
specification. That is, declarative knowledge about the word’s meaning and
information about its syntax and morphology which is necessary for
constructing the word’s syntactic environment. For example, the lemma for she
requires the word to be used of a female and that any following present tense
main verb must have the suffix -s attached to it for subject-verb agreement, the
lemma for like requires a subject that expresses the role of EXPERIENCER
and an object that expresses the role of THEME, and the lemma for give
152
Longxing Wei
requires a subject that expresses the role of AGENT, a direct object that
expresses the role of THEME and an indirect object that expresses the role of
BENEFACTIVE/GOAL. The elements required by a lemma appear in a
particular order as specified. The lemma for a particular word also contains
information about the word’s composition in terms of phonological segments
and its syllable and accent structure, and it may contain information about the
word’s register, the kind of discourse it typically enters into, and about its
pragmatics, stylistics, and affect. According to Levelt (1989), “It is in the
lemmas of the mental lexicon that conceptual information is linked to
grammatical function” (p. 162). In other words, lemmas are abstract entries in
the mental lexicon and underlie surface configurations of speech production.
Researchers differ in viewing how appropriate lemmas are activated to
convey the speaker’s intended meaning. For De Bot and Schreuder (1993),
lemmas are activated by pieces of conceptual structure sent by the Verbalizer.
For Poulisse and Bongaerts (1994), lemmas are not only activated by
conceptual information but also by a language cue sent by the Conceptualizer.
For Myers-Scotton and Jake (1995), lemmas are activated by language-specific
semantic/pragmatic feature bundles that come from the Conceptualizer.
According Myers-Scotton and Jake (2000), lexical access takes place on the
basis of the information contained in the speaker’s preverbal message. This is
because the speaker’s preverbal message in the Conceptualizer activates
language-specific semantic/pragmatic feature bundles at the interface between
the Conceptualizer and the mental lexicon. As Green (1993) assumes, a lemma
is activated if it matches part of the conceptual structure created by the
Conceptualizer. According to Grosjean (1997), whatever the origin and the
nature of information that lemmas may receive, the actual choice of lemmas
must be based on the information sent to the lexicon.
Although there is some disagreement about the nature of the lemma
representation in models of multi-layered levels of speech production (De Bot,
1992; Bock & Levelt, 1994; Kempen & Huijbers, 1983; Level, 1989, 1995;
Roelofs, 1992; Myers-Scotton, 1997, 2002; Myers-Scotton & Jake, 1995; Wei,
2001, 2002), it is generally assumed that lemmas are language-specific for
lexicalization patterns of a particular language (Talmy, 1985; Jackendoff,
1997). One often cited example is from Talmy (1985): (English) The bottle
floated into the cave. (Spanish) La botella entró a la cueva flotando. The
Intrasentential Codeswitching as Conceptual Projection of Lemmas in~
153
underlying structure for these sentences is proposed as [DO [MOVE
BOTTLE]: FIN [BOTTLE LOC IN CAVE]]: FLOAT BOTTLE (Bierwisch
and Schreuder, 1992). This shows that the lexicalization pattern differs across
the two languages. While English can ‘conflate’ MOVE with manner
condition FLOAT into a single lemma, Spanish cannot. Thus, in English a
semantic form in the mental lexicon is available for both motion and the
manner of motion to be expressed, but in Spanish a different pattern is
required. This paper assumes that language-specific lemmas form
interconnection between the lexical features and conceptual features, which
map to and from syntax (cf., ‘The distributed lexical/conceptual feature model’
by Kroll & De Groot, 1997). It further assumes that each lemma in the
bilingual mental lexicon is tagged for a specific language known and supports
the realization of an actual lexeme. In other words, language-specific lemmas
in the bilingual mental lexicon activate language-specific sets of
morphosyntactic procedures in the speech production Formulator.
In Levelt’s model (1989) semantic and syntactic information constitutes the
lemma of the lexical item while morphological and phonological information
constitutes the form of the lexical item. The conceptual information in the
preverbal message activates the appropriate lexical items during the
formulation of a message. Slightly different from Levelt’s model, this paper
proposes that it is the preverbal message which activates language-specific
lemmas in the bilingual mental lexicon, and such activated lemmas send
directions to the Formulator for projecting frame-building morphosyntactic and
phonological procedures. It emphasizes that it is the semantic/pragmatic
feature bundles selected by the Conceptualizer which trigger the appropriate
lemmas into activity before the Formulator has access to the lexical items
stored in the mental lexicon. It is assumed that language-specific lemma
activation plays a crucial role in bilingual speech production, especially in ICS.
3. Towards a bilingual lemma activation model
Levelt’s model of speech production was designed for describing the major
components and processes of monolingual production. This model has to be
adapted to account for bilingual speech behavior such as ICS. It becomes
necessary to explain how bilinguals manage to keep the two (or more)
154
Longxing Wei
Conceptual level:
CONCEPTUALIZER
Discourse mode
monolingual?
bilingual?
-intrasentential CS?
Preverbal message
-which language to be the ML?
-which semantic/pragmatic feature bundles
to be desired
Lemmal level:
THE BILINGUAL
MENTAL LEXICON
Language-specific
lemma activation
EL & ML lemma congruent?
-lexical-conceptual structure
-predicate-argument structure
-morphological realization
patterns
no
compromise
yes
Directions to the Formulator
Functional level:
FORMULATOR
Morphosyntactic encoding
Positional level:
ARTICULATOR
Morphophonological encoding
Surface forms:
word order &
phonetic string
Speech
comprehension
A bilingual lemma activation model (Adapted from Levelt, 1989; Myers-Scotton &
Jake, 2000; Wei, 2002)
languages apart, while at the same time they may switch between the two (or
more) languages at a certain point during a discourse. The bilingual’s mental
lexicon contains language-specific lemmas, and activation of such lemmas
mediates between the Conceptualizer and the Formulator. However, as
assumed in this paper, not all language-specific lemmas are activated equally,
and unequal activation of lemmas in the bilingual mental lexicon may result in
ICS. Crucially, lemma activation is treated as one of the levels of speech
production in a bilingual lemma activation model.
Intrasentential Codeswitching as Conceptual Projection of Lemmas in~
155
The model consists of four levels of speech production. At the conceptual
level the Conceptualizer generates messages by attending to the bilingual’s
communicative intention in terms of the discourse mode and preverbal
message to be desired. If the speaker chooses the monolingual mode, of course,
no codeswitching will occur; if the speaker chooses the bilingual mode (for
‘language modes’ see Grosjean, 1997), then he/she has to decide whether ICS
will be performed. If the speaker decides to perform ICS, the Conceptualizer
then generates preverbal message containing the speaker’s choice of the
language as the Matrix Language (Myers-Scotton, 1993) (commonly known as
the ‘base’ or ‘host’ language’) to be used in codeswitching and
semantic/pragmatic feature bundles to be desired for the speaker’s
communicative intention. The output of the Conceptualizer is a preverbal
message, which gives input to the bilingual mental lexicon at the lemma level.
It is at this level that language-specific lemmas are activated. In order for ICS
to occur, lemma congruence checking between the languages involved must
come into play. ‘Lemma congruence’ is defined as “a match between the ML
and the EL at the lemma level with respect to linguistically relevant features”
(Myers-Scotton & Jake, 1995, p. 985) (ML: Matrix Language; EL: Embedded
Language, commonly known as the ‘guest’ language). Lemma congruence
checking takes place at three levels: at the level of lexical-conceptual structure,
at the level of predicate-argument structure, and at the level of morphological
realization patterns. Lemma congruence checking is regarded as an organizing
principle for ICS (Wei, 2000b, 2002). If the ML lemma and the EL lemma are
congruent at each of these levels, the speaker can proceed with the bilingual
mode for ICS. If the ML lemma and the EL lemma are incongruent at any of
these levels, the speaker has to give up the bilingual mode and go back to the
monolingual mode instead. If this happens, no ICS will occur. Only when there
is a match between the ML and the EL at the lemma level or compromise
strategies are taken, directions will be sent to the Formulator at the functional
level for morphosyntactic encoding by observing a set of structural principles
governing ICS. The successfully encoded morphosyntactic material will then
be sent to the Articulator at the positional level for morphophonological
encoding. The successfully encoded morphophonological material will then be
the speech output. In other words, surface forms including word order and
phonetic string will be produced for speech comprehension.
156
Longxing Wei
4. Differential activation of participating languages in ICS
Two of the most important distinctions have been made in the ICS literature:
the distinction between content and system morphemes and the distinction
between the ML and the EL. The first distinction determines what types of
morphemes can be switched, and the second distinction determines how
switched morphemes can occur in a sentence. It is these distinctions which
constrain ICS configurations.
4.1. Content vs. system morphemes and lemma projection
One of the oppositions structuring ICS utterances is the distinction between
content and system morphemes (Myers-Scotton, 1993 [1997]). This distinction
depends on two lexical features: [± thematic role assigner/receiver] and [±
quantification], and is theoretically independent of the lexical categorization
feature of the morpheme. Nouns, descriptive adjectives, most verbs, and some
prepositions are prototypical content morphemes. They can be characterized as
categories that are potential thematic role assigners or receivers. Those
morphemes that assign or receive thematic roles are content morphemes. In
contrast to content morphemes, system morphemes do not assign or receive a
thematic role of any kind. Prototypical system morphemes are quantifiers,
specifiers and inflectional affixes.
In this paper, the content vs. system morpheme distinction is slightly recast
in light of the sources of morphemes. Assumptions regarding the sources of
morphemes refer to differences in the levels of abstract lexical structure.
According to Myers-Scotton and Jake (1995) and Wei (1996, 2000a, 2002), at
the conceptual level, speaker intentions are mapped onto semantic/pragmatic
feature bundles. While speaker intentions are not initially language-specific,
these bundles are language-specific. This is the mapping of speaker intentions
to lemma entries. It is assumed that abstract entries in the mental lexicon
support the surface realization of actual lexemes. Lemmas contain all aspects
of lexical information necessary to project a morphosyntactic frame. This is
because they activate morphosyntactic procedures spelling out the lexical
knowledge of the lexical entry. Such ‘directly-elected’ (Bock & Levelt, 1994;
Wei, 1996; Myers-Scotton, 2002) morphemes supported by those activated
Intrasentential Codeswitching as Conceptual Projection of Lemmas in~
157
lemmas in the mental lexicon are content morphemes.
According to Wei (1996, 2000a/b), some morphemes are selected from a
limited number of choices in the projection of the structure required by other
lexical items. Such morphemes are ‘indirectly-elected’ along with ‘directlyelected’ elements. Indirectly-elected morphemes are required as part of the
realization of the predicate-argument structure and morphological realization
patterns, but they do not represent lexical concepts independent of the directlyelected elements with which they are accessed. Indirectly-elected elements
include prepositions like to in listen to the story and at in look at the picture,
and particles like up in pick up the key and on in put on the coat. The treatment
of indirect election followed here relates speaker intentions to lemma entries.
For example, in certain semantic/pragmatic contexts, such as ‘definiteness,’
‘specificity,’ etc., the lemma supporting a noun directly elected by a nominal
argument projects the phrase structure with a determiner, such as a possessive,
a demonstrative or an article. This means that the possessive, the demonstrative
or the article is indirectly elected by the functional level projections of the
lemma supporting the noun via activated semantic/pragmatic features. In other
words, indirectly-elected lemmas only arise from projections of directlyelected lemmas (e.g., my brother read the book on that table.). The lexicalconceptual structure of a particular lemma entry of a noun requires a
determiner to serve certain semantic/pragmatic purposes.
According to Wei (2000a/b), certain morphemes are the result of the spelling
out of aspects of the morphological realization of a lexeme. They do not reflect
the semantic/pragmatic features conflated in the lexical-conceptual structure
but are only activated in the projection of morphophonological encoding at the
positional level. For example, case assignment reflects predicate-argument
structure and some phi-features license phonetic realizations, but they are not
indirectly elected by semantic/pragmatic features as required by speaker
intentions. Such morphemes are called ‘structurally-assigned’ morphemes
(Wei, 1996).
In addition to the distinction between content and system morphemes, the 4M model of Myers-Scotton and Jake (2000) categorizes three types of system
morphemes. It categorizes indirectly-elected morphemes as ‘early’ system
morphemes. They are called ‘early’ in the sense that they are activated at the
lemma level together with the lexical heads (e.g., the prepositions in listen to
158
Longxing Wei
and look at, and the particles in pick up and put on). Also, with a definite
referent, the determiner is indirectly elected as an early system morpheme by
the lemma underlying the noun in English, with a specifying or particularizing
effect. This is because the feature of an indirectly-elected morpheme is part of
the conceptual structure activated by the speaker’s intention. What makes the
4-M model different from the earlier model (Myers-Scotton, 1993) is that it
categorizes structurally-assigned system morphemes as ‘late’ system
morphemes, and it further subcategorizes late system morphemes into two
types: ‘bridges’ and ‘outsiders.’
According to the 4-M model, unlike an early system morpheme, a ‘bridge’
system morpheme is not activated at the conceptual level with a content
morpheme, but rather integrates a content morpheme into a larger constituent.
For example, the genitive/possessive of as in the property of the church is a
bridge system morpheme. The preposition of is structurally required in the
English grammar to integrate the [NP] the church into the [NP] the property to
realize the possessive case. Similarly, while the preposition of in I’ve never
thought of that is an early system morpheme, the preposition of in students of
linguistics is a late bridge system morpheme. A late system morpheme is
called a ‘bridge’ in the sense that it depends on information within the maximal
project of a category in which it occurs.
Unlike a bridge system morpheme, an ‘outsider’ system morpheme is
structurally assigned at the positional level (i.e., required in the spelling out of
aspects of the morphological realization patterns). For example, 3rd person
present tense -s and auxiliary verbs like do, have, be, and will/shall (future) are
outsider system morphemes. The 4-M model includes as outsider system
morphemes any morpheme under INFL which cannot be realized without
coindexing with another element in the sentence. Accordingly, tense marking
in English also belongs to this category. A late system morpheme is called an
‘outsider’ in the sense that it depends on grammatical information outside its
own maximal projection. That is, while bridge system morphemes refer not to
the head of the maximal projection in which they occur, but to other
information in this maximal projection, outsider system morphemes refer to
information outside their maximal projection of the head.
Relevant to the current discussion is that the 4-M model categorizes
morphemes into four types based on how they are elected. While information
Intrasentential Codeswitching as Conceptual Projection of Lemmas in~
159
about all types of morphemes is present in lemmas, information about content
morphemes and early system morpheme is salient at the conceptual level, and
information about late system morphemes becomes salient at the positional
level of the Formulator.
The content vs. system morpheme distinction underlies all linguistic
systems, and ICS is no exception. Based on their different sources, the division
of morphemes into those which are directly-elected and those which are not
results in two types of morphemes: content morphemes and system
morphemes. Directly-elected morphemes, those supported by lemma entries
activated by the Conceptualizer, are content morphemes; morphemes whose
lexical entries are ‘called’ by the Formulator in the mapping of lexicalconceptual structure onto predicate-argument structure or predicate-argument
structure onto morphophonological realization are system morphemes. Both
indirectly-elected and structurally-assigned morphemes are system
morphemes, but their sources are different. This difference lies in the fact that
some system morphemes represent lexical-conceptual structure, but others do
not.
4.2. Matrix vs. embedded languages and sentential frame projection
The other opposition structuring ICS utterances is the distinction between
the ML and the EL (Myers-Scotton, 1993 [1997]). The ML vs. EL distinction
is much more than a heuristic device for labeling constituents. Rather, it
underlies the differential participation of the ML and the EL in shaping ICS
utterances, with the ML having the dominant role. This dominance is most
dramatic in ML + EL constituents: only the ML can project the sentential
‘frame’ (i.e., the morpheme order, the inflections and other functional items).
The assumption underlying this model is that the languages participating in
ICS have unequal roles. One language is more central than the other in
sentential frame-building. The ML is more activated than the EL in a discourse
involving ICS and the occurrence of its morphemes is more frequent and freer
than that of the EL. The ML is the language which the speakers engaged in
ICS will identify as the ‘main language’ being used.
It is the speaker’s preverbal message at the conceptual level that determines
which language to be used as the ML if the speaker chooses the bilingual mode
and ICS. The ML chosen at the conceptual level, together with the
160
Longxing Wei
semantic/pragmatic feature bundles as desired, activates language-specific
lemmas at the lemma level for sending directions to the Formulator at the
functional level (see the figure). The ML vs. EL distinction is important in
analyzing and explaining bilingual production involving ICS.
4.3. Structural principles governing ICS
Three structural principles governing ICS are essential in explaining
grammatical constraints on ICS and predicting possible ML + EL constituents
and their structural configurations. In the Matrix Language Frame (MLF)
model (Myers-Scotton, 1993 [1997]), two principles are proposed under the
Matrix Language Hypothesis: the system morpheme principle and the
morpheme order principle. The system morpheme principle specifies that in
mixed constituents, all ‘syntactically relevant’ system morphemes come only
from the ML. The EL only supplies content morphemes to the sentential frame
projected by the ML. The morpheme order principle specifies that in mixed
constituents, surface morpheme order must not violate that of the ML. In
discussing compromise strategies in ICS (Wei, 2001, 2002), the lemma
congruence checking principle is formulated (cf. Myers-Scotton & Jake,
1995). This principle specifies that lemmas activated from the EL must be
congruent with the ML counterparts in order for ICS to occur; otherwise,
compromise strategies must be taken in order for ICS to be possible (see the
following section). With the content vs. system morpheme distinction and the
ML vs. EL distinction, these structural principles govern ICS. These principles
are further tested in this study of the nature of the bilingual mental lexicon in
relation to bilingual production with special reference to ICS involving several
language pairs.
5. Unequal activation of lemmas in ICS
The bilingual mental lexicon includes lemmas from more than one
language. It is true that there is a universal set of semantic/pragmatic feature
bundles that are available for the lexical-conceptual structuring of lemmas, but
lemmas are most likely language-specific because of cross-linguistic
incongruence or pragmatic considerations. Consequently, semantic or
pragmatic mismatches between the two languages may result.
Intrasentential Codeswitching as Conceptual Projection of Lemmas in~
161
This paper investigates ICS patterns on the assumption that it is the unequal
activation of lemmas in the bilingual mental lexicon that drives and constrains
ICS structural configurations. It focuses on language-specific
semantic/pragmatic feature bundles, morphosyntactic procedures, and the
lemma congruence checking procedures and compromise strategies. Hereby,
the three structural principles governing ICS are tested.
The type of codeswitching discussed in this paper is called “classic
codeswitching” in Myers-Scotton (2002: 8), which specifically refers to speech
containing switched items performed by the bilingual speakers who are
proficient enough to produce well-formed utterances in each of the languages
participating in codeswitching. In addition, such speakers are proficient enough
in using either participating language as the ML (i.e., the abstract
morphosyntactic frame of the utterances containing switched items). Similarly,
Muysken (2000) proposes a particular type of codeswitching (one of the three
patterns of intra-sentential code-mixing (his term)) called “insertion.” What he
may imply by the term “congruent lexicalization of material from different
lexical inventories into a shared grammatical structure” (2000: 3) is that the
features of two languages must be congruent in various ways in order for
insertion to occur. Muysken cites some examples from codeswitching
involving creoles, such as Sranan with Dutch, or the examples from
codeswitching between the two closely related languages, such as Frisian and
Dutch. Based on such examples, he agues that “congruent lexicalization often
involves bidirectional code-mixing, since there is no dominant matrix
language” (2000: 132). Although those examples are not from classic
codeswitching as defined by Myers-Scotton (2002), they offer insights into
other types of codeswitching and other codeswitching data that the MLF
model or the 4-M model alone cannot fully account for. The investigation of
examples not from classic codeswitching is beyond the scope of this paper.
The ICS data were collected from natural conversations in various settings
for a comprehensive research project on multilingual production involving
ICS. In this paper, only some typical instances of ICS are discussed. The
bilinguals for the study include those whose native language is Chinese and
speak English as a second language and those whose native language is
Japanese and speak English as a second language. All those bilinguals were
undergraduate or graduate students in the United States. The ICS instances
162
Longxing Wei
were tape-recorded, transcribed and analyzed by those who knew and spoke
the languages.
5.1. Language-specific lemma activation for content morphemes
The figure shows that it is at the conceptual level that, if the speaker chooses
a bilingual mode and ICS, his/her preverbal message determines which
semantic/pragmatic feature bundle to be selected as desired before lemmas are
activated at the lemma level. In other words, it is the speaker’s communicative
intention that motivates the activation of language-specific lemmas in his/her
bilingual mental lexicon. As mentioned and emphasized earlier, each lemma in
the bilingual mental lexicon is tagged for a specific language and supports the
realization of an actual lexeme. This paper further suggests that one of the
major reasons why content and system morphemes are accessed differently lies
in the fact that it is content morphemes, rather than system morphemes, that
contain semantic/pragmatic feature bundles. The speaker may choose certain
EL content morphemes at a certain point during a discourse to realize his/her
intended or desired meaning in his/her preverbal message. It is in this sense
that content and system morphemes are not equally activated, and it is also in
this sense that language-specific lemmas are conceptually projected in ICS.
Below are some typical ICS examples.
(Abbreviations in glosses for ICS examples: AFFIRM = affirmative; AUX =
auxiliary; CL = classifier of noun; COP = copula; EMPH = emphatic; INTJ =
interjection; LOC = location; NOM = nominative; OBJ = object; PART =
particle; PAST = past; PERF = perfect; POSS = possessive; PREP =
preposition; PROG = progressive; TOP = topic.)
(Chinese/English)
(1) wo you liang-fen paper mingtian bixu jiaoshangqu, ke wo
I have two-CL
tomorrow must turn in
but I
xianzai
yi-fen hai mei finish ne.
at the moment one-CL yet not
AFFIRM
“I must turn in two papers tomorrow, but at the moment I haven’t
finished one yet.”
Intrasentential Codeswitching as Conceptual Projection of Lemmas in~
163
(2) wo summer bu take course le.
I
not
AFFIRM
“I won’t take any course in summer.”
(3) you xuduo homework yao zhuo; hai
you hao ji-pian article
have a lot of
must do in addition have quite a few-CL
xiang qu library-de
computer shang check yixia.
want go
-POSS
on
once
“I’ve a lot of homework to do. In addition, I’ve quite a few articles I
want to go to check on the library’s computer.”
(4) naxie visiting scholar bu
shi hen youqian ma,
those
not/EMPH COP very rich
AFFIRM
bi
women student you qian duo le.
PREP/than us
have money more AFFIRM
“Aren’t those visiting scholars very rich? They have a lot more money
than us students.”
(5) mei you zijide jiqi
feichang bu fanbian feichang inconvenient.
not have own machine very
not convenient very
“It’s very inconvenient if I don’t have my own machine, very
inconvenient.”
In (1) paper and finish are EL content morphemes, but the EL system
morphemes ‘-s’ for plural marking and ‘-ed’ for perfect aspect marking do not
appear. In (2) summer and take course are EL content morphemes. It should be
noticed that the Chinese similar preposition ‘zai’ (‘in’) is optional in realizing
an adverbial of time or place. Also, while English requires an auxiliary verb, a
system morpheme, to negate a predicate verb, Chinese does not have one for
this grammatical purpose. ‘bu’ (‘not’), a system morpheme, comes from the
ML. In (3) homework, article, library, computer and check are EL content
morphemes, but the system morphemes ‘xuduo’ (‘much/a lot of’), ‘ji-pain’ (‘a
few’-CL) and ‘-de’ (adjective marker) come from the ML. The EL infinitive
marker ‘to’ does not appear. In (4) visiting scholar and student are EL content
morphemes, but the demonstrative pronoun ‘naxie’ (‘those’) in an ML system
164
Longxing Wei
morpheme, and there is no EL plural morpheme ‘-s’ on scholar or student. In
(5) inconvenient is an EL content morpheme, but ‘feichang’ (‘very’) is an ML
system morpheme.
The above examples show that EL content morphemes behave differently
from EL system morphemes in ICS. All the system morphemes come only
from the ML. Although Chinese does not possess inflectional morphemes for
number or tense/aspect marking, even if the EL morphemes appear, they do
not carry the system morphemes required for the grammatical purposes. The
similar phenomena are also observed in the Japanese/English ICS instances.
(Japanese/English)
(6) dore
gurai koko ni
stay suru no?
how long about here LOC
do QUE
“About how long will you stay here?”
(7) ima wa
summer course o
tot-te
iru
now TOP
OBJ take-PROG AUX/be
n.
PART
“I’m taking summer courses now.”
(8) ii
desu keredomo tuition ga
totemo expensive desu.
good COP/be but
NOM very
COP/be
“It’s good, but the tuition is very expensive.”
(9) Nihon demo saikin kekko drug o ya-tte
iru
hito
Japan also recently rather
OBJ do-PROG AUX/be people
ga
ooi yo.
PART/NOM many INTJ
“Recently in Japan people who are doing drugs are also many.”
(10) muzukashikat-ta to
iu ka, aa
sore essay datt-ta
difficult-PAST PART say PART PART that
PART-PAST
kara
wakara-nai,
um.
because Understand-not PART
Intrasentential Codeswitching as Conceptual Projection of Lemmas in~
165
“It could be said to be difficult, mm because I didn’t understand that essay,
mm.”
In (6) stay is an EL content morpheme used in conjunction with the
ML ‘suru’ (‘do’). In (7) summer course is an EL content morpheme phrase,
but ‘o’ is an ML system morpheme marking the accusative case. In (8) tuition
and expensive are EL content morphemes, but the EL definite article ‘the’ does
not appear before tuition, and ‘totemo’ (‘very’), a system morpheme, comes
from the ML. In (9) drug is an EL content morpheme, but the EL plural ‘-s’, a
system morpheme, does not appear. In (10) essay is an EL content morpheme,
but ‘sore’ (‘that’), a system morpheme, comes from the ML.
The above ICS instances provide empirical evidence that EL content and
system morphemes are not equally activated. EL content morphemes are free
to occur if they are activated to meet the speaker’s communicative intention at
a certain point during a discourse, but EL system morphemes are not. It also
should be noticed that although Chinese and Japanese have a different basic
word order and Japanese possesses morphology for case and tense/aspect
marking, but Chinese does not, the system morpheme principle applies to both
language pairs involved in ICS.
5.2. Language-specific lemma activation for morphosyntactic
procedures
As assumed, the ML and the EL participating in ICS play unequal roles.
Once one language is chosen by the speaker as the ML for the bilingual mode
at the conceptual level, it will be more activated than the EL. This is because it
is the ML which builds the sentential frame for ML + EL constituents and
provides all the system morphemes and most of the content morphemes. The
figure shows that the activated language-specific lemmas send directions to the
Formulator at the functional level for morphosyntactic encoding. In other
words, at the conceptual level the discourse mode is chosen with one language
as the ML and then corresponding language-specific lemmas are activated at
the lemma level to realize the speaker’s preverbal message, resulting in
language-specific morphosyntax. Bilingual speakers may choose any of the
languages they know as the ML based on several factors, such as the languages
known to the interlocutors, the speech settings, the conversation topics, the
social motivations, and so on. The study of these various factors in choosing
166
Longxing Wei
one language rather than the other as the ML is beyond the scope of this paper.
What is emphasized here is that whichever language that is activated as the ML
will play a central role in sentential frame-building throughout the discourse
involving ICS.
The above Chinese/English ICS instances provide evidence for the
morphosyntactic procedures controlled by the ML. Chinese and English share
the same basic V-O order, but Chinese is very flexible in the arrangement of
sentential elements. Once Chinese is chosen as the ML, it sets up the sentential
frame (i.e., morphosyntactic procedure) into which the EL content morphemes
are switched. In (1) the adverbial of time ‘mingtian’ (‘tomorrow’) goes before
the predicate and the direct object goes before the verbs ‘jiaoshangqu’ (‘turn
in’) and finish respectively. In (2) the adverbial of time (in) summer follows the
subject rather than being placed in the sentence initial or final position as in
English. In (3) the adverbial of place ‘library-de computer shang’ (‘on the
library’s computer’) goes before the predicate verb check. In (4) the
prepositional phrase for comparison ‘bi women student’ (‘than us student’)
goes before the predicate. These are just some of the common Chinese
arrangements of sentential elements, but they are not permitted in English. The
Japanese/English ICS instances provide further evidence. Unlike English,
Japanese has the O-V order, and the verb final order is always maintained once
Japanese is chosen as the ML. In (6) the adverbial of place ‘koko ni’ (‘in here’)
goes before the verb ‘stay suru’. In (7) the object summer course occurs before
the verb. In (8) the copula ‘desu’ (‘be’) follows the predicate adjective ‘totemo
(‘very’) expensive’. Such a Japanese morphosyntactic procedure is also
observed in (9) and (10).
De Bot and Schreuder (1993) note bilingual speakers are able to separate
different language systems and to mix them in a bilingual mode. According to
Grosjean (1989, 1997), the source of variability in the amount of language
mixing is the language mode the bilingual speaker is in: monolingual,
bilingual, or anywhere else on the continuum between these two extremes.
This paper proposes that knowing about a particular position on the languagemode continuum is not sufficient enough for knowing about how and why this
particular position is adopted. This paper assumes that the bilingual’s adoption
of a particular position on the language-mode continuum is closely associated
with particular levels of activation of the ML and the EL. Bilingual speakers
Intrasentential Codeswitching as Conceptual Projection of Lemmas in~
167
also know that the ML and the EL play unequal roles in ICS. The so-called
‘mixed’ speech is actually governed by a set of underlying structural principles.
Below are more examples of such a language separation.
(English/Chinese)
(11) It’s not easy for students to get jiangxuejin. Only youxiude students
scholarship
excellent
can get it.
“It’s not easy for students to get scholarship. Only excellent students
can get it.”
(12) If I buy a used car, I’ll buy a used Riben che. ershou
Riben
Japanese car second-hand Japanese
che are laokaode duo
and much cheaper.
car
reliable a lot more
“If I buy a used car, I’ll buy a used Japanese car. Second-hand Japanese
cars are a lot more reliable and much cheaper.”
(13) Some graduate students fabiao wenzhang at conferences, but it’s
present papers
difficult to get papers published.
“Some graduate students present papers at conferences, but it’s difficult
to get papers published.”
(11)-(13) show that the Chinese speakers may switch to English as the ML
with Chinese as the EL. In (11) the infinitive clause with the formal subject
pronoun ‘it’ is a typical English construction, but Chinese does not possess a
similar one. In (12) both the main clause and the embedded clause are from the
ML. In (13) the predicate fabiao wenzhang (‘present papers’) goes before the
adverbial of place (‘at conferences’), which is an unacceptable word order in
Chinese.
(English/Japanese)
(14) It’s totemo muzukashi to find a convenient and yasui apartment here.
very difficult
cheap
168
Longxing Wei
“It’s very difficult to find a convenient and cheap apartment here.”
(15) supa
is close from here, but I have to walk juugo fun
gurai
supermarket
fifteen minutes about
to the bus stop.
“The supermarket is close from here, but I have to walk about fifteen
minutes to the bus stop.”
In (14) the infinitive clause with the formal subject pronoun ‘it’ is not
available in Japanese. In (15) the speaker uses the English verb initial
construction, rather than the verb final construction typical in Japanese.
Grosjean (1989, 1997) suggest that the bilingual’s language system is
organized in two subsets that can be activated and deactivated independently or
simultaneously, each to a particular degree. This paper suggests that such an
activation or deactivation occurs at the conceptual level of the speech
production process. It is at the conceptual level that the speaker decides
whether to adopt a monolingual or bilingual mode. If the bilingual mode is
chosen, the speaker will choose between intersentential or intrasentential CS. If
ICS is chosen, the speaker must decide which language to be the ML and
which semantic/pragmatic feature bundles to be desired. Such decisions are
made by the speaker’s preverbal message at the conceptual level. If the
bilingual mode and ICS are chosen, language-specific semantic/pragmatic
feature bundles activate language-specific lemmas at the lemma level. The
activated lemmas in turn send directions to the Formulator at the functional
level for the projection of language-specific morphosyntactic procedures. The
projected language-specific morphosyntactic procedures are
morphophonologically realized at the positional level (i.e., surface forms).
5.3. Lemma congruence checking procedures and compromise
strategies
Lemma congruence between languages is “a match between the ML and the
EL at the lemma level with respect to linguistically relevant features” (MyersScotton & Jake, 1995, p. 985). This paper further proposes that lemma
congruence checking is an organizing principle for possible ICS
configurations, and congruence checking must take place at the level of
Intrasentential Codeswitching as Conceptual Projection of Lemmas in~
169
lexical-conceptual structure, at the level of predicate-argument structure, and at
the level of morphological realization patterns. Relevant to the current study
are the first two levels.
5.3.1. Lemma congruence checking at the level of lexical-conceptual
structure
The figure shows that at the conceptual level speakers do not produce
surface morphemes but rather make appropriate choices about the
semantic/pragmatic information that they intend to convey. The
semantic/pragmatic feature bundles chosen at the conceptual level activate the
lemmas in the mental lexicon which will support surface morphemes, and the
activated EL lemmas must be congruent with the ML lemmas in terms of their
lexical conceptual structure, predicate-argument structure, and morphological
realization patterns in order for ICS to occur. Most ICS instances studied in this
paper show sufficient congruence between the EL and the ML lemmas which
support the existing lexemes in both languages.
Bilinguals may switch to another language either intentionally or
unintentionally and either intersententially or intrasententially, in the sense that
codeswitching becomes part of some bilinguals’ daily practice or natural
speech patterns (cf. Li, 1996). However, the language pairs involved may differ
in lexical-conceptual structure. If such a difference is only partial, there is still
sufficient cross-linguistic congruence. Grosjean (1982) report that some codeswitches are motivated by the lack of a particular word in one of the languages
or by the greater availability of a word in the other language. Such switches are
deliberate or intentional. This paper suggests that a partial difference at the
level of lexical-conceptual structure is one of the major reasons why certain
morphemes from another language are switched.
(Chinese/English)
(16) wo you liang-fen paper mingtian bixu jiaoshangqu.
I have two-CL
tomorrow must turn in
“I must turn in two papers tomorrow.”
(17) wo xiawu qu jian wode advisor.
I afternoon go see my
170
Longxing Wei
“I’m going to see my advisor this afternoon.”
(18) zhu zai
zheli hen fanbian, meitian you school bus.
live PREP/LOC here very convenient everyday have
“It’s very convenient to live here since there is a school bus everyday.”
In (16) a paper in English may mean any written piece of work, such as an
article, an essay, a composition, but the Chinese equivalent lexeme ‘zhi’
(‘paper’) itself only means a piece of paper to wrap things up in or to write
something on. In (17) an advisor in English means a professor or instructor
who offers advice or counsel to students regarding their academic progress,
improvement, course requirements and sequential arrangements, thesis or
dissertation writing, research in progress, and so on. Most English advisors are
also those who will recommend their students to professional agencies.
However, a Chinese ‘daoshi’ (‘advisor’) does not assume the same
responsibilities as an English advisor. His/her only responsibility is to guide
his/her students in writing research papers, theses or dissertations. In (18) an
English school bus means a bus which transports students to and from a
school. Although Chinese has the equivalent lexeme, ‘xiaoche’ (‘school bus’),
it only means a bus which transports a school’s sports or performance team or
equipment.
(Japanese/English)
(19) futatsu no
bedroom ga
ate, hitori,
Maria to
two POSS
NOM COP one person
and
iu ko
wa
hitori
de one bedroom o
call person TOP one person PREP/by
OBJ
mot-te
imasu yo.
Have-PROG AUX PART/AFFIRM
“We have two bedrooms. One person, called Maria, has one bedroom.”
(20) anata wa
registration o shimashi-ta ka?
you TOP
OBJ do-PERF QUE
“Have you done your registration?”
Intrasentential Codeswitching as Conceptual Projection of Lemmas in~
171
In (19) the possible reason for the speaker to switch to bedroom is that in
Japan the concept of ‘bedroom’ is relatively new. A traditional Japanese room
is used not only for sleeping but also for eating, studying, meeting guests, or
other daily family activities. In (20) the speaker switches to registration for the
possible reason that in Japanese colleges/universities, although students must
register for the courses to take, they are not free to select the courses which
they are interested in and want to take. The English lexeme may carry a special
meaning for the speaker to express his intended meaning more accurately.
As observed, bilingual speakers choose particular lexical items from another
language at a certain point during a discourse because of such cross-linguistic
differences in lexical-conceptual structure. This paper assumes that although
any language is capable of expressing its speakers’ semantic/pragmatic
intentions, lexicalization patterns are language-specific. As observed by Li
(1996), Nishimura (1997), and Wei (2001, 2002), in may cases exact
translation across languages is inaccurate or incomplete. Because language
cues may have different values, the speaker may switch to an EL lexical item
to realize his/her intended meaning. When the language cue specifies a
particular language at a certain point during a discourse involving ICS, the
lexical item from that language receives activation. This means that conceptual
information and language cues must work together in activating languagespecific lemmas in the mental lexicon for the speaker’s communicative
intention.
5.3.2. Lemma congruence checking at the level of predicate-argument
structure
In ICS it is the ML which controls the predicate-argument structure because
it supplies system morphemes and subcategorization frames for verbs.
Although morphosyntactic procedures are realized by the Formulator at the
functional level, before morphosyntactic directions are sent to the Formulator,
lemmas from both languages can be activated at a certain point during a
discourse. Thus, lemma congruence checking at the level of lexical-conceptual
structure alone is not sufficient enough for ICS to occur. Lemma congruence at
the level of predicate-argument structure must also come into play.
As observed, the speakers involved in Chinese/English ICS tend to use
many EL verbs and verb phrases as well as EL nouns and noun phrases. One
172
Longxing Wei
of the reasons for this is that the EL predicate-argument structure can easily fit
into the syntactic slots provided by the ML.
(Chinese/English)
(21) wode che you give me trouble le.
my car again
PART/PERF
“My car has given me trouble again.”
(22) ni dei xiang bangfa make money.
you must think way
“You must think of ways to make money.”
(23) ta gang dao, ta dei xue drive.
he just arrive he must learn
“He just arrived, and he must learn how to drive.”
In (21) the subcategorization frame for the verb give is congruent with that
for the equivalent verb in Chinese. In (22) make money fits the Chinese
morphosyntactic frame. In (23) drive in its infinitive form is used as the object
of the main verb ‘xue’ (‘learn’) is also morphosyntactically allowed in
Chinese. However, in Japanese/English ICS instances there are very few
English verbs or verb phrases switched into the Japanese morphosyntactic
frame. Instead, when the speakers switch between Japanese and English, they
choose other English lexical items, such as nouns/noun phrases and
adjectives/adjective phrases. The possible reason for this is that even though
the predicate-argument structure may be congruent between Japanese and
English, the Japanese morphosyntactic frame may reject English verbs or verb
phrases.
5.3.3. Lemma incongruence and compromise strategies
One of the major reasons why languages differ is that different languages
may lexicalize concepts differently. In other words, lexical representations are
language-specific. If the bilingual mode is chosen at the conceptual level, but
the lemmas activated from the EL do not sufficiently match the ML
counterparts, some compromise strategies must be taken in order for ICS to
Intrasentential Codeswitching as Conceptual Projection of Lemmas in~
173
occur. One of the compromise strategies is the production of EL islands
(Myers-Scotton, 1997; Jake & Myers-Scotton, 1997; Wei, 2002). Otherwise,
only intersentential codeswitching will be possible if the speaker still wants to
adopt a bilingual mode. An EL island is a constituent consisting of an EL
content morpheme with only other EL morphemes, including EL system
morphemes. Such a compromise strategy can be taken at the level of lexicalconceptual structure or at the level of predicate-argument structure.
(Chinese/English)
(24) ni neng-bu-neng give me a ride?
you can-not-can
“Can you give me a ride?”
(25) name ni mingtian call me.
then you tomorrow
“Then you call me tomorrow.”
(26) na wo yi dian come to pick you up.
so I one o’clock
“So, I’ll come to pick you up at one o’clock.”
In (24) while in the EL the lexical-conceptual structure of the means of
transportation is conflated in the noun ride as the direct object of the verb, in
the ML it is conflated in the verb ‘song’ (‘send’), but the verb itself may not
contain any means of transportation. The speaker chooses the EL expression
probably because he wants to be more specific than he can be with the Chinese
verb. Thus, when the EL lemma is activated, the whole VP is produced as an
EL island. In (25) since the speaker chooses the EL lemma underlying call me,
the EL lexical-conceptual structure is activated and the whole VP is accessed
as an EL island. While in the EL the semantic features of ‘communicate with
by telephone’ are conflated in the verb call, in the ML equivalent ‘da dianhua
gei wo’ (‘make phone to me’), the same meaning is realized by both the verb
and its direct object as well as its indirect object. In (26) the speaker chooses
the EL lemma underlying come to pick you up, and thus the VP with a
pronominal object before the particle satellite up is accessed. The infinitive to,
174
Longxing Wei
an EL system morpheme, also appears in the island. The speaker prefers pick
up because this phrasal verb means ‘to take on as a passenger’, but the Chinese
equivalent verb ‘jei’ usually does not. Such cross-linguistic differences in
lexical-conceptual structure are also shown in the Japanese/English ICS
instances. For example,
(Japanese/English)
(27) nan to
iu n
desu
ka, Amerika
what PART say NOM COP/be QUE America
de Christian toka ga
ooi desh-oo.
dakara nanka
PREP/LOC so on PART/NOM many COP/be-AFFIRM so something
before marriage sex ga
dame
mitai da.
NOM prohibited like COP/be
“Whatever you say, in America there’re many Christians and other
such people, so things like before marriage sex is prohibited, or the
like.”
In (27) the speaker is talking about sex before marriage in the American
context. He switches to the EL lexical item sex and also PP before marriage,
an EL island, where the EL word order is observed. The speaker chooses the
EL expression probably because most Japanese still keep their traditional
concept of sex before marriage. In America sex before marriage is not
necessarily bounded with marriage, but in Japan it is usually, if not always, so.
In other cases, although there is a sufficient congruence between the lexicalconceptual structures across the languages involved in ICS, the predicateargument structures may differ. If such incongruence occurs, but the
semantic/pragmatic feature bundles desired by the speaker activate the EL
lemma for his/her communicative intention, a radical compromise strategy
must be taken in order for the EL material to be accessed. In other words, the
activated EL lemma must send special directions to the Formulator for the EL
material to be realized in an EL island.
(Chinese/English)
(28) ta jingchang bangzhu wo with my computer work.
he often
help
me
Intrasentential Codeswitching as Conceptual Projection of Lemmas in~
175
“He often helps me with my computer work.”
(29) ta jingchang fails students in exams.
she often
“She often fails students in exams.”
(30) ni biye
hou keyi teach English to non-native speakers.
you graduate after can
“After you graduate, you can teach English to non-native speakers.”
In (31) the THEME my computer work is introduced by the preposition with
in the EL, but in the ML any THEME must be introduced by the verb itself.
The speaker’s choice of the EL material is realized as the EL island because of
the incongruence between the EL and the ML in predicate-argument structure.
In (32) the VP headed by fail is an EL island, with all the system morphemes
from the EL. In the EL fail is used as a causative verb and this takes the
grammatical subject as the AGENT who makes the failure happen. The
Chinese equivalent lexeme to ‘fail’ is ‘shibai,’ but it is used only as a
noncausative verb and takes the grammatical subject as the EXPERIENCER.
The speaker prefers the EL concept, but there is incongruence between the EL
and the ML in predicate-argument structure, the result is the production of an
EL island. In (33) while in the EL the RECIPIENT in the VP can be
introduced by a preposition like to in the indirect object dative construction, the
Chinese equivalent VP headed by ‘jiao’ (‘teach’) only permits the double
object dative construction. Again, since the speaker prefers the EL material, but
the ML does not accept the mapping which the EL PP would project at the
level of predicate-argument structure, the result is the production of an EL
island.
6. Conclusion
The naturally occurring ICS instances studied in this paper offer several
implications for exploring the nature of the bilingual mental lexicon and the
bilingual speech production process and structural principles.
First, the bilingual mental lexicon contains lemmas rather than lexemes from
176
Longxing Wei
the languages known, and these lemmas are tagged for a specific language. If
the speaker chooses the bilingual mode and is engaged in ICS, he/she may
activate the language-specific lemmas as desired from his/her bilingual mental
lexicon. Language-specific lemmas contain information about the word’s
meaning, semantics, pragmatics, syntax, and morphology, and such
information is necessary for using the word appropriately and for constructing
its syntactic environment.
Second, the bilingual speech production process contains the same levels as
those contained in the monolingual speech production process. However, at the
conceptual level, the bilingual speaker makes several choices about the
language mode, monolingual or bilingual, to be used and semantic/pragmatic
feature bundles to convey his/her communicative intention. If the speaker
chooses the bilingual mode, he/she still has to choose between intersentential
or intrasentential CS. If the speaker activates language-specific lemmas in
his/her bilingual mental lexicon for particular lexemes, ICS comes into play.
Third, the bilingual’s languages are turned ‘on’ all the time during a
discourse involving ICS, but these languages are never activated equally at the
same time. One language is more activated as the ML than the other as the EL.
It is the speaker who chooses whichever language as the ML. It is only the ML
which controls morphosyntactic procedures and provides both content and
system morphemes at a much higher frequency. The EL only supplies content
morphemes as desired by the speaker to be switched into the ML sentential
frame. This is because only conceptually activated EL lemmas in the bilingual
mental lexicon can project EL content morphemes for ICS.
Fourth, the bilingual can activate lemmas from whatever language as the EL
during a discourse involving ICS, but these activated EL lemmas must be
sufficiently congruent with the ML counterparts at the level of lexicalconceptual structure and at the level of predicate-argument structure (also at the
level of morphological realization patterns). If lemma incongruence occurs
between the languages involved in ICS at any of these levels, radical
compromise strategies must be taken in order for the EL material to be realized
in ICS configurations.
Intrasentential Codeswitching as Conceptual Projection of Lemmas in~
177
Acknowledgements
I would like to thank the anonymous reviewer for his very insightful and
invaluable comments on an earlier version of this paper. Without those
comments, the improvement of this paper would have been impossible. The
paper was much improved, though probably not as much as the reviewer
would have wished. I would also like to thank Hyeongdong Park, the assistant
editor of Journal of Cognitive Science, for his great assistance and patience.
References
Bierwisch, M., & Schreuder, R. (1992). From concepts to lexical items. Cognition,
42, 23-60.
Bock, K., & Levelt, W. (1994). Language production: Grammatical encoding. In M.
A. Gernsbacher (Ed.), Handbook of psycholinguistics (pp. 945-984). New York:
Academic Press.
De Bot, K. (1992). A bilingual production model: Levelt’s ‘speaking’ model
adapted. Applied Linguistics, 13, 1-24.
De Bot, K., & Schreuder, R. (1993). Word production and the bilingual lexicon. In
R. Schreuder, & B. Weltens (Eds.), The bilingual lexicon (pp. 191-214).
Amsterdam: John Benjamins.
Grosjean, F. (1982). Life with two languages: An introduction to bilingualism.
Cambridge, MA: Harvard University Press.
Grosjean, F. (1989). Neurolinguistics, beware! The bilingual is not two monolinguals
in one person. Brain and Language, 36, 3-15.
Grosjean, F. (1997). Processing mixed languages: Issues, findings, and models. In A.
M. B. De Groot, & J. F. Kroll (Eds.), Tutorials in bilingualism: Psycholinguistic
perspectives (pp. 225-254). Mahwah, NJ: Erlbaum
Jackendoff, R. (1997). The architecture of the language faculty. Cambridge, MA:
MIT Press.
Jake, J. L., & Myers-Scotton, C. (1997). Codeswitching and compromise strategies:
Implications for lexical structure. International Journal of Bilingualism, 1, 2539.
Kempen, D., & Huijbers, P. (1983). The lexicalization process in sentence
production and naming: Indirect election of words. Cognition, 14, 185-209.
Kroll, J. F., & De Groot, A. M. B. (1997). Lexical and conceptual memory in the
bilingual: Mapping form to meaning in two languages. In A. M. B. De Groot, &
J. F. Kroll (Eds.), Tutorials in bilingualism: Psycholinguistic perspectives (pp.
178
Longxing Wei
169-199). Mahwah, NJ: Erlbaum.
Levelt, W. (1989). Speaking: From intention to articulation. Cambridge, MA: MIT
Press.
Levelt, W. (1995). The ability to speak: From intentions to spoken words. European
Review, 3, 13-23.
Li, D. C. S. (1996). Issues in bilingualism and biculturalism: A Hong Kong case
study. New York: Peter Lang.
Muysken, P. (2000). Bilingual speech: A typology of code-mixing. Cambridge
University Press.
Myers-Scotton, C. (1993, [1997]). Duelling languages: Grammatical structure in
codeswitching. Oxford, UK: Clarendon Press.
Myers-Scotton, C. (2002). Contact linguistics: Bilingual encounters and
grammatical outcomes. New York: Oxford University Press.
Myers-Scotton, C., & Jake, J. L. (1995). Matching lemmas in a bilingual language
competence and production model: Evidence from intrasentential code
switching. Linguistics, 33, 981-1024.
Myers-Scotton, C., & Jake, J. L. (2000). Explaining aspects of codeswitching and
their implications. In J. Nicol (Ed.), One mind, two languages: Bilingual
language processing (pp. 91-125). Oxford, UK: Blackwell.
Nishimura, M. (1997). Japanese/English code-switching: Syntax and pragmatics.
New York: Peter Lang.
Poulisse, N., & Bongaerts, T. (1994). First language use in second language
production. Applied Linguistics, 15, 36-57.
Roelofs, A. (1992). A spreading activation theory of lemma retrieval in speaking.
Cognition, 42, 107-142.
Talmy, L. (1985). Lexicalization patterns: Semantic structures in lexical form. In T.
Shopen (Ed.), Language typology and syntactic description III: Grammatical
categories and the lexicon (pp. 51-149). New York: Cambridge University
Press.
Wei, L. (1996). Organizing principles behind codeswitching and interlanguage
development in early adult second language acquisition. In J. Arnold, R. Blake,
B. Davidson, S. Schwenter, & J. Solomon (Eds.), Sociolinguistic variation:
Data, theory, and analysis. Selected papers from NWAV 23 (pp. 417-431).
Stanford, CA: Center for the Study of Language and Information.
Wei, L. (2000a). Unequal election of morphemes in adult second language
acquisition. Applied Linguistics, 21, (1), 106-140.
Wei, L. (2000b). Types of morphemes and their implications for second language
morpheme acquisition. International Journal of Bilingualism, 4, (1), 29-43.
Wei, L. (2001). Lemma congruence checking between languages as an organizing
Intrasentential Codeswitching as Conceptual Projection of Lemmas in~
179
principle in intrasentential codeswitching. International Journal of Bilingualism,
5, 153-173.
Wei, L. (2002). The bilingual mental lexicon and speech production process. Brain
and Language, 81, 691-707.
Title of the paper
Intrasentential Codeswitching as Conceptual Projection of Lemmas
in the Bilingual Mental Lexicon
Mailing address
Dr. Longxing Wei
Linguistics Department
Montclair State University
Upper Montclair, NJ 07043
USA
[email protected]