Like in English and como,como que,and like in Spanish in the speech of Southern Arizona bilinguals

Abstract

Aims and Objectives/Purpose/Research Questions:

This study analyzes the use of like in English and como, como que, and like in Spanish in the speech of bilinguals from Southern Arizona to assess the possible influence of like in English on its equivalents in Spanish in a language contact situation in which English is the majority language.

Design/Methodology/Approach:

Drawing on a discourse-pragmatic variationist approach, this study analyzes the use of like in English and its Spanish equivalents in recorded conversations between nine pairs of young Spanish-English bilingual friends from Southern Arizona.

Data and Analysis:

3389 tokens of like in English and its Spanish equivalents from 18 hours of recorded conversations (9 hours in each language) were analyzed quantitatively. The analysis assesses the relative frequencies of these variants and their syntactic positioning as clause-external discourse markers and clause-internal discourse particles. The independent variables of the analysis were the language of the conversation and the sex and language dominance of the participants.

Findings/Conclusions:

Contact with English did not appear to radically influence the use of como, como que, and like in Spanish in the speech of these bilinguals. In the speech of the same bilinguals, like in English was much more frequent and occurred in many more syntactic positions than its Spanish equivalents.

Originality:

This is the first study of discourse-pragmatic features in contact to analyze the use of discourse markers and discourse particles in both the donor and the recipient language in the speech of the same bilinguals.

Significance/Implications:

These results contribute to our knowledge of the limited interaction of linguistic repertoires in the speech of bilinguals at the discourse level even in language contact situations with a majority language. They also underline the ability of bilinguals to both understand and reproduce the subtleties of the use of these features in the two languages they speak.

Keywords

Discourse-pragmatic features bilinguals language contact like

Introduction

Thirty years ago, an innovative use of like as a discourse marker was observed in American English (Schourup, 1985; Underhill, 1988). Like has now been documented not only in American English, but in other English-speaking countries, including Canada (D’Arcy, 2005, 2017), England (Andersen, 2001), and Australia (Miller, 2009). The use of como in Spanish as a discourse marker equivalent to like appears to have been a recent development, found in the speech of both Spanish-English bilinguals (Erker & Bruso, 2017; Jimenez & Flores-Ferrán, 2018; Kern, 2014; Said-Mohand, 2008; Sánchez Muñoz, 2007) and Spanish monolinguals (Jørgensen & Stenström, 2009). In addition to como, the use of como que (Jimenez & Flores-Ferrán, 2018; Jørgensen & Stenström, 2009; Kern, 2014; Mata, 2016) and like (Kern, 2014; Mata, 2016) have been observed in Spanish as equivalents of like in English. Equivalents of like have also been documented in other languages, including comme (Sankoff et al., 1997) and genre (Fleischman & Yaguello, 2004) in French, ke’ilu in Hebrew (Maschler, 2009), and kao in Croatian (Hlavac, 2006).

Discourse markers are defined by Schiffrin (1987) as “sequentially dependent elements that bracket units of talk” (p. 31). In her seminal work on like, D’Arcy (2015, 2017) distinguishes between the use of like as a discourse marker outside of the clause and its use as a discourse particle within it. The following examples of the use of like in English (1) and its three equivalents, como, como que, and like, in Spanish (2) as clause-external discourse markers (bolded) and clause-internal discourse particles (underlined) were extracted from informal conversations in English and Spanish between pairs of Spanish-English bilingual friends from Southern Arizona.

(1) And you can like, you can have like some kind of censor that’s connected to like, um, some kind of sensor that measures like voltage for example (7B2).

(2) No, como que estaba cansado y como que me siento que, como que, no, um, no estoy haciendo on purpose, like… Es como que ya quiero empezar a hacer más cosas, pero como que, no puedo, como que, no, y mi muffin top está como igual (8B1).

“No, like I was tired and like I feel that, like, no, um, I’m not doing it on purpose, like… It’s like, I now want to do more things, but like, I can’t, like, no, and my muffin top is like the same.”

This study analyzes the use of like in English and the use of como, como que, and like in Spanish as discourse markers and discourse particles in the speech of the same bilinguals from Southern Arizona and the similarities and differences in their use within and between languages. Since English represents the majority language in this community, it is hypothesized that contact with like in English will influence the use of its Spanish equivalents. The next section reviews previous studies of discourse-pragmatic features in contact situations. This is followed by a discussion of the grammaticalization of like in English and its Spanish equivalents as discourse markers and discourse particles. Finally, after reviewing the methodology of the study, the analysis of these variants in English and Spanish in the speech of the same bilinguals is presented and discussed. The results of this study shed light on the interaction of linguistic repertoires in the speech of bilinguals at the discourse level in language contact situations in which one language is culturally dominant.

Discourse-pragmatic features in contact situations

Discourse markers and discourse particles, such as like and como, are examples of discourse-pragmatic features. According to Pichler (2013), discourse-pragmatic features are “syntactically optional elements” that “make little or no contribution to the truth-conditional meaning of their host units” and perform functions “to express speaker stance; to guide utterance interpretation; and to structure discourse” (p. 4). In addition to discourse markers and discourse particles, Andersen (2014) adds that discourse-pragmatic features include interjections, greetings, expletives, vocatives, general extenders, tag questions, quotatives, focus markers, intonation, and paralinguistic features, such as gesture (p. 23).

Many studies have analyzed the borrowing of discourse-pragmatic features in language contact situations (Aaron, 2004; Andersen, 2014; Brody, 1987; 1995; Flores-Ferrán, 2014; Hlavac, 2006; Matras, 1998; 2000; Muysken, 2000; Peterson & Vaattovaara, 2014; Salmons, 1990; Torres, 2002; Treffers-Daller, 1994; Van Hout & Muysken, 1994). The borrowing of discourse-pragmatic features appears to lend support to the “anything goes” hypothesis set forth by Thomason and Kaufman (1988). These authors propose that the degree of contact between languages determines their convergence. Casual contact leads to the borrowing of lexical items. More intense contact leads to the borrowing of non-lexical items. Cases of extreme contact lead to syntactic, morphological, and phonological changes (Thomason & Kaufman, 1988, pp. 74–75).

The transfer of the discursive or pragmatic functions of discourse-pragmatic features without form has also been documented by several studies. These studies include analyses of the equivalents of like in English in several other languages (Hlavac, 2006; Kern, 2014; Sankoff et al., 1997). Zavala (2001) also found the transfer of the discourse functions of –mi in Quechua to pues in Spanish among Spanish-Quechua bilinguals in Peru.

When specifically considering the borrowing of discourse markers in contact situations, several studies have gone further to examine the cognitive and functional motivations of their transfer. The first of these motivations for the convergence of discourse markers is a lessening the cognitive load (Maschler, 2009; Matras, 2000). Maschler (2009) proposes that there is a cognitive realm in which discourse markers can refer to cognitive processes taking place in the mind during verbalization. Matras (2000) argues that the cognitive motivation of using bilingual discourse markers is so strong that it can “override the social and communicative constraints on the discourse,” which may explain unintentional choices or slips (pp. 514–515). Although this cognitive motivation has not been explored for like, Lipski (2005) proposes that the insertion of English discourse markers including so and you know in Spanish discourse may be automatic and against the preference of Spanish-English bilinguals, signaling “a metalevel in which discourse is framed in terms of English” (p. 12). A second motivation for convergence is a filling of the gap (Hill & Hill, 1986; Hlavac, 2006). Hill and Hill (1986) identified function words from Spanish including como that had been incorporated in the syncretic language of Mexicano in Mexico. In a study of Croatian-English bilinguals, Hlavac (2006) proposes that the discourse marker that fulfills the most functions is selected in the process of convergence. Hlavac (2006) argues that kao in Croatian is used more frequently than its equivalent like in English because it fulfills more functions. A third motivation for the convergence of discourse markers is a marking or strengthening of discourse boundaries (Dajko & Carmichael, 2014; de Rooij, 2000; Maschler, 1994, 1997, 2009). Maschler (2009) argues that Hebrew-English bilinguals use English discourse markers including like in Hebrew discourse “to metalanguage” discourse boundaries, that is, to mark Hebrew discourse strategically through English. The choice of English discourse markers instead of their Hebrew counterparts is iconic, fulfilling the same discourse function as the discourse markers themselves and the use of English discourse markers is often accompanied by pauses to preserve maximal contrast. Similarly, de Rooij (2000) and Dajko and Carmichael (2014) observe that bilingual discourse markers are often accompanied by pauses and prosodic cues to increase their saliency. The following section turns to the use of like and como as discourse markers and discourse particles.

The grammaticalization of like and como as discourse markers and discourse particles

Discourse markers and discourse particles, such as like and como, are believed to originate from lexical equivalents through a process of grammaticalization (Brinton, 2006; Traugott, 1995). While the grammaticalization of like in English has been explored in several studies (D’Arcy, 2005, 2017; Romaine & Lange, 1991), much less is known about the development of its Spanish equivalents. In a previous study of the use of like and como as discourse markers and discourse particles in Southern Arizona, Kern (2014) suggests that como may be following a similar grammaticalization path proposed for like in English, reflecting an internal change in Spanish that may have been triggered by English contact.

Regarding the development of like in English, Romaine and Lange (1991) propose that the use of like as a discourse marker developed on the pathway of preposition > conjunction > discourse marker (p. 261). Citing a lack of evidence of the direct development of the discourse marker like from its use as a conjunction, D’Arcy (2017) argues that the discourse marker like developed instead from its use as a sentence adverb and proposes the pathway of preposition > conjunction > adverb > discourse marker (p. 60). In addition to decategorialization in the development of like as a discourse marker, the author observes that like also exhibits other characteristics of grammaticalization, including semantic bleaching, pragmatic strengthening through its use to fulfill pragmatic functions, and divergence through its multiple uses as a preposition, conjunction, and discourse marker (p. 65).

Crucial to hypotheses of the grammaticalization of like, D’Arcy (2005; 2017) further distinguishes between the use of like as a discourse marker that appears clause-externally (3a) and as a discourse particle that appears clause-internally (3b).

(3) a. Like when they were about to take her in for surgery, like I was

born (15B2).

b. It was like the first cool party we went to (1B2).

As a clause-external discourse marker, like can modify both a matrix clause and a subordinate clause (D’Arcy, 2017; Underhill, 1988) and can appear on the left or right periphery. As a clause-internal discourse particle, like can precede or enter noun phrases, verb phrases, adjectival phrases, adverbial phrases, or prepositional phrases (Andersen, 2001; D’Arcy, 2017).

D’Arcy (2005) observes that the clause-internal use of like as a discourse particle does not align with traditional characteristics of grammaticalization because its scope has narrowed instead of broadened (p. 72). By analyzing the use of like as a discourse marker and discourse particle in apparent time, she proposes that the use of like as a discourse particle developed from its use as a discourse marker in the opposite direction, incorporating itself back into the syntax of the sentence and narrowing its scope (pp. 218–219). D’Arcy (2017) returns to this hypothesis, affirms the pathway of conjunction > adverb > discourse marker > discourse particle for the development of like, and argues that the grammaticalization of like is ongoing (p. 65). Her hypothesis of the development of like as a discourse marker and discourse particle is illustrated in Figure 1.

Figure 1.

The grammaticalization of like in English.

Perhaps como in Spanish is following a similar grammaticalization path proposed for like in English as a discourse marker and discourse particle (Kern, 2014). Discursive uses of como in Spanish have been documented in both Spanish-English bilingual communities Erker & Bruso, 2017; (Kern, 2014; Said-Mohand, 2008; Sánchez Muñoz, 2007) and monolingual Spanish communities (Jørgensen & Stenström, 2009). Similar to like, como can also be used both clause-externally as a discourse marker (4a) and clause-internally as a discourse particle (4b) (Kern, 2014).

(4) a. Como, no sé, no sé cómo funciona la verdad (1B1).

‘Like, I don’t know, I don’t know how it works to tell you the truth.’

b. Hemos tenido que escribir como ensayitos (12B2).

“We have had to write like short essays.”

Figure 2 uses D’Arcy’s (2017) pathway of preposition > conjunction > adverb > discourse marker > discourse particle to illustrate a possible parallel grammaticalization path of como in Spanish.

Figure 2.

The grammaticalization of como in Spanish.

In addition to como, two other variants in Spanish have been documented in previous studies as equivalents to like in English. Firstly, the expression como que has been found to fulfill similar discursive functions to like in English in the speech of Spanish-English bilinguals (Jimenez & Flores-Ferrán, 2018; Kern, 2014; Mata, 2016) and monolingual Spanish speakers (Jørgensen & Stenström, 2009). Jimenez and Flores-Ferrán (2018) conclude that como que is a “free variant” of como; although como que is less frequent than como, it has not been replaced (p. 165). With the exception of Jimenez and Flores-Ferrán (2018), previous studies have either grouped como and como que together in one category or analyzed one or the other. The use of como que as a discourse marker (5a) and as a discourse particle (5b) has not been addressed in previous studies, but is documented in this study.

(5) a. ¿Como que empezó a gritar? (1B1).

“Like she began to scream?”

b. Estaba pensando como que en literature (12B2).

“I was thinking like about literature.”

Secondly, the insertion of English like in Spanish discourse (6) has also been found in Spanish-English bilingual communities (Kern, 2014; Mata, 2016) and represents a third possible variant in Spanish as an equivalent to like in English.

(6) La otra vez le robaron algo, no sé, like, pues, no sé, hace como dos días o tres (8B2).

“The other time they stole something from him, I don’t know, like, well, I don’t know, like two or three days ago.”

The present study seeks to assess if D’Arcy’s (2017) hypothesis for the development of like is consistent with the use of como, como que, and like in Spanish. If like in English is influencing the use of the Spanish variants, it is hypothesized that the distribution and syntactic positioning of like and its Spanish variants as discourse markers and discourse particles would be similar in the speech of the same bilinguals. It would also be expected that English-dominant bilinguals would use como, como que, and like in Spanish more frequently than Spanish-dominant bilinguals. The following section turns to extralinguistic factors in the use of like in English and the Spanish variants as discourse markers and discourse particles.

Extralinguistic factors in the use of like and its Spanish equivalents as discourse markers and discourse particles

Several extralinguistic factors have been explored in the use of like and its Spanish equivalents as discourse markers and discourse particles. Younger speakers are viewed as using like more frequently than older speakers (Dailey-O’Cain, 2000); however, older speakers also use like, and the syntactic positions in which like can appear as a discourse marker and as a discourse particle are found in all age groups (D’Arcy, 2017). With the exception of Mata (2016), who concluded that como que was a core filler across age groups, the use of como has primarily been documented among young people (Jørgensen and Stenström, 2009; Kern, 2014; Said-Mohand, 2008; Sánchez Muñoz, 2007).

Regarding sex, women are viewed as using like more frequently than men (Dailey-O’Cain, 2000), but there are conflicting results according to its actual use, with some studies finding this to be the case (Andersen, 2001; Tagliamonte, 2005) and others concluding the opposite, that men use like more than women (Dailey-O’Cain, 2000). D’Arcy (2017) did not find an overall proportional difference in the use of like according to sex, but she concluded that women were the leaders of linguistic change in the use of like as a discourse marker, while men were the leaders in the use of like as a discourse particle. For como, Kern (2014) concluded that young female bilinguals use como significantly more frequently than young male bilinguals, while Said-Mohand (2008) only found sex to be significant for quotative constructions with como, with young women using them more than young men.

Other extralinguistic factors that have been found to be significant for the use of como and its Spanish variants, including register, generation, variety, and Spanish proficiency. In Southern California, Sánchez Muñoz (2007) concluded that como was conditioned by register because it was used significantly more frequently in informal conversations than in formal interviews and class presentations in an academic setting. In his study of the use of fillers according to generation and community on the San Diego–Tijuana border, Mata (2016) found that como que was used by both first- and second-generation speakers in both San Diego and Tijuana, but like in Spanish was used only by second-generation, Spanish heritage speakers in San Diego. In Florida, Said-Mohand (2008) did not find a significant difference in the frequency of como between second- and third-generation Spanish-English bilinguals or according to Spanish proficiency. Finally, in a study of como and como que, Jimenez and Flores-Ferrán (2018) found that speakers of Mexican, Puerto Rican, and Colombian varieties favored como, while speakers of Dominican, Cuban, and Honduran varieties favored como que.

Controlling for both age and variety, the present study analyzes the use of like and its equivalent Spanish variants among young second- and third-generation Spanish heritage speakers of Mexican descent who were born and raised in Southern Arizona. It includes the social factors of sex and language dominance. Although Said-Mohand (2008) previously analyzed como according to Spanish proficiency and did not find it to be significant, language dominance has not been previously explored for the use of like in English or the Spanish variants among Spanish-English bilinguals. The following section turns to the research questions of the present study.

Research questions

In order to investigate the possible influence of like in English on its Spanish equivalents in the speech of bilinguals in a language contact situation in which English is the dominant language, the following research questions are addressed.

What are the similarities and differences in the relative frequencies of the use of like in English and como, como que, and like in Spanish as discourse markers and discourse particles?

How do like in English and como, como que, and like in Spanish compare in terms of the syntactic positions in which they can appear and their distribution as discourse markers and discourse particles?

Is sex significant in the relative frequencies and/or the syntactic positioning and distribution of like in English and como, como que, and like in Spanish as discourse markers and discourse particles?

Is language dominance significant in the relative frequencies and/or the syntactic positioning and distribution of like in English and como, como que, and like in Spanish as discourse markers and discourse particles?

The novelty of the present study is that it analyzes the use of like in English and its Spanish equivalents in the speech of the same Spanish-English bilinguals. In doing so, the interaction of these variants in the linguistic repertoires of the same bilinguals can be assessed.

Methods of data collection

In order to eliminate the potential for linguistic accommodation to the presence of an investigator (Rickford & McNair-Knox, 1994), conversations between friends who already knew each other were recorded without an investigator present. This method of data collection has been shown to elicit an informal register with a higher frequency of discourse-pragmatic features than more formal interviews (Sánchez Muñoz, 2007). Nevertheless, the observer’s paradox (Labov, 1972) is still in play because there are recording devices present during the interaction, and participants are aware that they are being recorded.

Nine pairs of Spanish-English bilingual friends from Southern Arizona were audio and video recorded while conversing with each other for 1 hour without the presence of the investigator on two occasions that were one week apart: one occasion in Spanish and the other in English. Half of the pairs of participants had their first conversation in English and their second conversation in Spanish, and the other half of the pairs of participants completed the reverse. On each occasion, participants were instructed to speak as they would normally speak when conversing with their friend in each language. The participants self-selected their partners, and each pair of participants had previously known each other for at least three months. If they wanted or needed a topic to sustain their conversation, participants were provided with optional conversation questions in English or Spanish that were typed on a slip of paper and compiled in an envelope. This method of data collection provided 18 hours of recorded conversations between pairs of Spanish-English bilingual friends, 9 hours in each language. The conversations have an informal register demonstrated through overlapping turns, expletives, code-switching, and the use of discourse-pragmatic features.

The 18 participants in the study were second- or third-generation Spanish-English bilinguals between 18 and 24 years old who were of Mexican descent and born and raised in Southern Arizona. Considering Arizona’s geographic location north of Mexico, there is a large population of Hispanic or Latino origin, and Spanish is spoken in many homes. Although Spanish has been maintained in the community through the constant immigration of new Spanish speakers, English is the majority language, and there is a generational shift toward English monolingualism (Jaramillo, 1995). All of the participants were Spanish heritage language learners who were recruited from upper-division undergraduate courses in Spanish at a large public university in Southern Arizona. Although definitions of heritage speakers can be broad and narrow, all heritage language learners come into the classroom with some degree of personal experience with the heritage language and are bilingual in various degrees (Valdés, 2000).

The participants completed a demographic information form, adapted from Carvalho (2012–) and a Bilingual Language Profile (Birdsong, Gertken, & Amengual, 2012). Regarding the social factors included in the study, the participants were equally divided by sex. One pair of participants consisted of two female friends, one pair of participants consisted of two male friends, and the remaining seven pairs of participants were male–female pairs. To assess language dominance in this study, a dominance score was calculated for each participant using the Bilingual Language Profile adapted from Birdsong et al. (2012) by Carvalho (2012–). These scores have a range from –218 to +218, with a score near zero corresponding to balanced bilingualism, negative values indicating Spanish dominance, and positive values indicating English dominance. The range of dominance scores for the participants was from –60.57 to 49.952, with an average score of 3.20. Male participants had an average dominance score of 5.06 and female participants had an average dominance score of 1.34. Although language dominance scores reflected a broad range for inclusion as a factor in the quantitative analysis, the average participant was a balanced bilingual.

Methods of data analysis

Unlike morphosyntactic and phonological variables, discourse-pragmatic features are optional, do not occur in a fixed context, and are multifunctional, often fulfilling different, sometimes overlapping, functions in different contexts (Pichler, 2010, p. 588). These characteristics complicate their analysis, and different methodologies of quantitative analysis have hampered cross-corpora comparability. Discourse-pragmatic features have been analyzed quantitatively either by counting all of the instances of a single form while disregarding function or by counting all of the forms that fulfill a single function (Tagliamonte, 2012). With the notable exception of D’Arcy (2005, 2017), who follows the first approach to analyze the syntactic positioning of like, the second approach has been most commonly followed in discourse-pragmatic variation.

The frequency of discourse-pragmatic features may be the most important aspect of their variation (Macaulay, 2002), and this presents more issues regarding the quantitative analysis of discourse-pragmatic features. Relative frequencies of discourse-pragmatic features according to word count per 1000 or 10,000 words are highly dependent on the methods of data collection (sociolinguistic interviews, conversations, etc.), which may favor certain discourse-pragmatic features over others (Macaulay, 2002), as well as how data is transcribed, such as including hesitations and false starts or not including them (Pichler, 2010). The analysis of frequency also does not take into account the contexts in which discourse-pragmatic features appear or do not appear (Pichler, 2010), violating Labov’s (1972) principle of accountability. D’Arcy (2005, 2017) takes into account Labov’s (1972) principle of accountability by identifying the contexts in which like appears and assessing when it appears and when it does not, but she does not take into account function.

The analysis of the use of like in English and como, como que, and like in Spanish in the speech of the bilinguals in this study includes 3389 tokens from 18 hours of recorded conversations, 9 hours in each language. Since both like and its Spanish equivalents can be used to fulfill lexical functions, a criterion was established to differentiate their use as discourse markers and discourse particles from their lexical equivalents. Following Fuller (2003), if discourse markers are removed from an utterance, both grammaticality and the semantic relationship between elements are maintained (Fuller, 2003, p. 186). The same criterion was applied to discourse particles. Examples of like in English that were used as a verb (7), adverb (8), preposition (9), or a conjunction (10) were eliminated because their removal violates grammaticality. These lexical uses of like were also identified and eliminated by D’Arcy (2015; 2017).

(7) I like this one (3B1).

(8) Well everything always goes like that (1B2).

(9) It looks like a Wal-Mart backpack (8B1).

(10) I feel like I’m in a hotel (15B1).

Examples of like that fulfilled a lexical function of exemplification (11) were eliminated in this study because their removal violates grammaticality. D’Arcy (2015, 2017) includes examples of exemplification in her analysis.

(11) I really love doing things with my hands like painting (12B1).

In addition, following D’Arcy (2005, 2017) and differing from Andersen (2001), examples of like that fulfilled a lexical function of approximation (12) were eliminated because the semantic relationship between elements is no longer intact upon their removal.

(12) She works like four hours per day, or maybe five (3B2).

Quotative constructions with like (13) and the referentially empty collocation it’s like (14) were also eliminated in this study.

(13) I was like, “You went to college?” and she was like, “For a semester” (17B2).

(14) It’s like, some of her music’s ok (15B2).

The remaining examples of like that, upon removal, did violate grammaticality or the semantic relationship between elements were included in the study and classified as discourse markers or discourse particles according to their syntactic positioning outside or inside of the clause, respectively. Any occurrences of like that appeared within a clause, but before or after a pause, that is, outside of the scope of the sentence, were classified as discourse markers, as shown in (15).

(15) I’m too stubborn to like, you know, get persuaded too easily (12B2).

The use of como, como que, and like in Spanish were similarly analyzed. Any token that violated grammaticality or the semantic relationship between elements in an utterance was removed from the study. Similar to the analysis of like in English, the removed tokens of como, como que, and like in Spanish included tokens that were used as adverbs, prepositions, or conjunctions, as well as tokens that fulfilled lexical functions of exemplification and approximation or appeared in quotative constructions. Any occurrences of the Spanish variants that appeared within a clause but before or after a pause were classified as discourse markers. Tokens of like that were inserted in Spanish discourse were included in the study, but tokens of like that were embedded in English code-switches within Spanish discourse were eliminated from the analysis.

Since lexical examples of like in English and como, como que, and like in Spanish, including those that fulfilled functions of exemplification and approximation or that appeared in quotative constructions, were removed from the analysis, the remaining tokens included in the analysis can be circumscribed as fulfilling the same discourse function of structuring units of discourse. Following Kern (2014), the present study refers to this discourse function of both like in English and the three Spanish variants as a focus discourse function. Kern (2014) borrows this term from Underhill (1988), who claims that like fulfills a focus discourse function, which marks “the most significant new information in a sentence—often, the point of the sentence” (p. 238). The focus discourse function performed by like in English and its Spanish equivalents in the present analysis does not refer to syntactic focus and does not necessarily mark new information. Instead, it refers to drawing focus to units of discourse, which can happen either broadly outside of the clause or narrowly within the clause. Consequently, the use of these variants as discourse markers outside of the clause and as discourse particles within the clause can be grouped together as performing the same discourse function. The difference between discourse markers and discourse particles is not a difference in function, but a difference in scope. A limitation of this analysis is glossing over the multifunctional characteristics of these features, as discussed by Andersen (2001), Miller and Weinert (1995), Schourup (1985), and Underhill (1988), among many others, for like, and Said-Mohand (2008) for como. While the present analysis does not assess micro-pragmatic functions, such as hedging (Underhill, 1988), the focus discourse function is sufficient for circumscribing like in English and its Spanish equivalents for quantitative analysis.

After identifying the variants to be included in the study, the frequencies of like in English and como, como que, and like in Spanish per 10,000 words were calculated for each participant to give an overall view of their use. These frequencies were calculated by dividing the total number of tokens of each variant for each participant by each participant’s overall word count in each language and multiplying by 10,000. Responding to Pichler’s (2010) call for the methodology of word counts to be explained in detail, whole word false starts were included in the word count, while non-word false starts were excluded. Planners (hmm, uh, ugh, um, ah), affirmations (uh-huh, ok), and interjections (ooh!, argh!) were included in the word count. Contractions were transcribed as one word, including wanna and gonna. Next, the overall distributions of like in English and the three Spanish variants as discourse markers and discourse particles were analyzed. Finally, the syntactic positioning of like in English and the three Spanish variants as discourse markers and discourse particles was explored and compared. Linear mixed-effects models were run in R (R Core Team) to assess the significance of language, language dominance, and sex in the frequency and distribution of the use of like in English and the Spanish variants as discourse markers and discourse particles. All of these models were run with random intercepts for each participant to consider variation among speakers.

Analysis

The following analysis assesses the use of like in English and its equivalent variants of como, como que, and like in Spanish. It begins by exploring the frequencies and overall distributions of the variants. This is followed by analyzing the overall distribution of the variants as discourse markers and discourse particles. Finally, the analysis zooms in to pinpoint the positioning of the variants as discourse markers and discourse particles.

Frequencies of the variants

The average frequencies of like in English and the three Spanish variants per 10,000 words per participant can be found in Table 1.

Table 1.

Average frequency of like in English and como, como que, and like in Spanish per 10,000 words per participant.

English	Spanish
like	como	como que	like
360.38	55.98	8.35	16.88

Figure 3 illustrates the average frequencies of like in English and the three Spanish variants per 10,000 words per participant according to participant, sex, and language dominance.

Figure 3.

Frequency of like in English and como, como que, and like in Spanish per 10,000 words according to participant, sex, and language dominance.

The participants’ use of like in English showed greater cross-individual variability and was much more frequent, an average of 360.38 times per 10,000 words, than their use of como (55.98 times per 10,000 words), como que (8.35 times per 10,000 words), and like (16.88 times per 10,000 words) in Spanish. Como is the preferred variant in Spanish, used more frequently than como que and like in Spanish. Linear mixed-effects models were carried out to assess the significance of language, language dominance, and sex in the frequency of the use of like in English and the Spanish variants. There was a main effect of language when comparing the frequency of like in English with como (χ²(1) = 41.17, p < 0.05), como que (χ²(1) = 50.07, p < 0.05), and like (χ²(1) = 48.34, p < 0.05) in Spanish, as well as all of the Spanish variants combined (χ²(1) = 37.39, p < 0.05). Sex and language dominance were not significant in any of the models. The high frequency of like in English does not appear to influence the frequency of its variants in Spanish, regardless of sex and language dominance. The following section turns to the overall distribution of like in English and the Spanish variants as discourse markers and discourse particles.

Overall distributions of the variants as discourse markers and discourse particles

The overall distribution of like in English and the three Spanish variants as discourse markers outside of the clause and discourse particles within the clause is presented in Table 2. Since discourse particles are considered to be a later development than discourse markers, analyzing the distribution of the use of like in English and como, como que, and like in Spanish as discourse markers and discourse particles can provide a snapshot of their development. Figure 4 illustrates the percent use of each variant as discourse particles according to participant, sex, and language dominance.

Table 2.

Distributions of like in English and como, como que, and like in Spanish as discourse markers (DMs) and discourse particles (DPs).

English		Spanish
like		como		como que		like
DM	DP	DM	DP	DM	DP	DM	DP
71.43% (2100/2940)	28.57% (840/2940)	72.01% (211/293)	27.99% (82/293)	90.32% (56/62)	9.68% (6/62)	100% (93/93)	0% (0/93)

Figure 4.

Percent use of like in English and como, como que, and like in Spanish as discourse particles according to participant, sex, and language dominance.

Illustrated in Table 2 and Figure 4, the overall distributions of like in English and como in Spanish as discourse markers and discourse particles appear to be conditioned similarly, with 28.57% (840/2940) and 27.99% (82/293) of tokens used as discourse particles. However, the percent use of como in Spanish as a discourse particle exhibits much more variability among participants, while the use of like does not. This result mirrors Kern (2014), who found that 22.73% (55/242) of tokens of como were used as discourse particles with high cross-individual variability. The use of como que and like in Spanish appear to be conditioned differently than like in English and como in Spanish, with only 9.68% (6/62) and 0% (0/93) of tokens used as discourse particles. The fact that like in Spanish is never used as a discourse particle distinguishes it from like in English and como and como que in Spanish.

Linear mixed-effects models were carried out to assess the significance of language, language dominance, and sex in the distribution of like in English and the Spanish variants as discourse markers and discourse particles, with the percent use of each variant as a discourse particle by each participant as the dependent variable. As expected, there was a main effect of language when comparing the percent use of like as a discourse particle in English with the percent use of como que (χ²(1) = 56.12, p < 0.05) and like (χ²(1) = 56.12, p < 0.05) as discourse particles in Spanish. Sex and language dominance were not significant. This result means that the percent use of like in English as a discourse particle is significantly greater than percent use of como que and like as discourse particles in Spanish, regardless of sex and language dominance. When comparing the percent use of like as a discourse particle in English and the percent use of como as a discourse particle in Spanish, there was a not a main effect of language; however, there was an interaction between language and language dominance, illustrated in Figure 5.

Figure 5.

Interaction between language and dominance in the percent use of like in English and como in Spanish as discourse particles.

Illustrated in Figure 5, there are no significant correlations in English, but in Spanish there is a correlation between language dominance and the use of como as a discourse particle for both male and female bilinguals. Specifically, as Spanish dominance increases (represented by the negative values in Figure 5) so too does the percent use of como as a discourse particle.

When comparing the percent use of like as a discourse particle in English and the combined Spanish variants as discourse particles, there were main effects of language (χ²(1) = 6.59, p < 0.05) and language dominance (χ²(1) = 5.05, p < 0.05) and an interaction between language and language dominance. The percent use of like in English as a discourse particle was significantly greater than the percent use of the combined Spanish variants as discourse particles. In Spanish, but not in English, the use of the combined Spanish variants as discourse particles is correlated with Spanish dominance (p < 0.5). The percent use of the variants as discourse particles increases as Spanish dominance increases (Figure 6).

Figure 6.

Interaction between language and dominance in the percent use of like in English and the combined Spanish variants as discourse particles.

In summary, like in English and como in Spanish are conditioned differently than como que and like in Spanish, according to their distribution as discourse markers and discourse particles. While like in English and como in Spanish are used as discourse particles clause-internally, the use of como que in Spanish as a discourse particle was very infrequent, and like in Spanish was never used as a discourse particle. Regarding the social factors of the study, sex was not significant, but the percent use of como in Spanish and all of the Spanish variants combined as discourse particles is positively correlated with Spanish dominance. As Spanish dominance increases, so too does the percent use of como and the combined Spanish variants as discourse particles. In the following section, the analysis continues by exploring the syntactic positioning of like in English and the three Spanish variants as discourse markers and discourse particles.

Positioning of the variants as discourse markers and discourse particles

The analysis of the syntactic positioning of like in English and the three Spanish variants as discourse markers and discourse particles provides a qualitative understanding of their use. Beginning with the use of these variants as discourse markers, the use of like in English as a discourse marker can occur clause-initially before matrix clauses (16) or subordinate clauses (17).

(16) Like when they were about to take her in for surgery, like I was born (15B2).

(17) Like if I get in an accident and something really bad happens to me and I go to the doctor, like they don’t cover those bills (16B2).

Following D’Arcy (2005), tokens of like in English that intervened between the head of the clause and the subject, as seen with clauses beginning with because (18) as well as in relative clauses (19), were found in the discourse of these bilinguals and classified as discourse markers that modified subordinate clauses.

(18) I felt the exclusion by them because like they would get together and they wouldn’t let me know (5B2).

(19) They’re not celebrities, but they’re the people who like act (15B2).

The use of like in English also appeared on the right periphery (20). Although these tokens are infrequent, they were also classified as discourse markers.

(20) Because it was like my family was one, and two, it was my friends and a lot of them didn’t speak Spanish you know, like (12B2).

Similar to like in English, the use of como in Spanish as a discourse marker was also found clause-initially before matrix clauses (21); however, unlike like, it does not appear before subordinate clauses and it does not appear on the right periphery.

(21) Y luego se queda, aquí, ((gesto)) como estoy bien flaco ahora pero aquí ((gesto)) estoy medio gordo (8B1).

“And then it stays, here, ((gesture)) like I’m pretty skinny now, but here ((gesture)) I’m a little fat.”

Similar to como in Spanish, the use of como que in Spanish as a discourse marker only appeared before matrix clauses (22) and did not appear on the right periphery.

(22) ¿Como que empezó a gritar? (1B1).

“Like she began to scream?”

Finally, all of the examples of like in Spanish occurred outside of the clause as discourse markers before matrix clauses (23), but four tokens appeared on the right periphery (24).

(23) La otra vez le robaron algo, no sé, like, pues, no sé, hace como dos días o tres (8B2).

“The last time they stole something, I don’t know, like, well, I don’t know, it was like two or three days ago.”

(24) Me siento que, como que, no um no estoy haciendo on purpose, like (8B1).

“I feel that, like, no, um, I’m not doing it on purpose, like.”

Table 3 summarizes the type of clause that each of these variants modify. All of the discourse marker variants highly favored positions before matrix clauses. Only like in English modified subordinated clauses but did so very infrequently (5.05%; 106/2100).

Table 3.

Type of clause of like in English and como, como que, and like in Spanish as discourse markers.

Type of clause	English	Spanish
	like	como	como que	like
Matrix	94.95% (1994/2100)	100% (211/211)	100% (56/56)	100% (93/93)
Subordinate	5.05% (106/2100)	0% (0/211)	0% (0/56)	0% (0/93)

Table 4 summarizes the positioning of the use of these variants as discourse markers on the left and right peripheries. All of the discourse marker variants favored the left periphery. Only the like variants in English and Spanish appeared on the right periphery, both very infrequently (1.67%; 35/2100 and 4.30%; 4/93, respectively). In summary, the differences in the use of like in English and como, como que, and like in Spanish as discourse markers are minimal, but it is important to note that none of the Spanish variants modified subordinate clauses, whereas like in English did appear before subordinate clauses, although very infrequently.

Table 4.

Periphery of like in English and como, como que, and like in Spanish as discourse markers.

Type of clause	English	Spanish
	like	como	como que	like
Left periphery	98.33% (2065/2100)	100% (211/211)	100% (56/56)	95.70% (89/93)
Right periphery	1.67% (35/2100)	0% (0/211)	0% (0/56)	4.30% (4/93)

The use of like in English as a discourse particle within a clause can modify a determiner phrase (25), a prepositional phrase (26), an adjectival phrase (27), or an adverbial phrase (28).

(25) It was like the first cool party we went to (1B2).

(26) I always see them like at XXX getting their makeup done (3B1).

(27) He told me internships there are like paid (5B1).

(28) I forget these things like really fast (12B2)

The use of like in English as a discourse particle can also appear embedded within noun phrases (29), prepositional phrases (30), and adjectival phrases (31).

(29) Get it on a little like cotton ball thing (1B2).

(30) She just came with like the manager (17B1).

(31) It was really like clean (12B2).

Moreover, the discourse particle like in English can occur in the verbal domain between the subject and the verb (32), between a modal and a verb (33) (including will (33a), would (33b), etc.), between an auxiliary and a participle (34a), between an auxiliary and a gerund (34b)), and between a verb + to + infinitive (35). There were also a few examples of like used in constructions with the passive voice (36).

(32) The beer pong table like was on the floor (5B2).

(33) a. If you’re using your phone they’ll like point at you (3B2).

b. I would like do my own thing (12B2).

(34) a. I’ve never like done that (5B2).

b. I was like walking around (12B1).

(35) I decided to like leave (12B1).

(36) I was like racially profiled (17B2).

The use of como in Spanish as a discourse particle within a clause can modify a determiner phrase (37), a prepositional phrase (38), an adjectival phrase (39), or an adverbial phrase (40).

(37) Hemos tenido que escribir como ensayitos (12B2).

“We have had to write like essays.”

(38) Está como en Phoenix (15B1).

“It is like in Phoenix.”

(39) Está como igual (8B1).

“It is like the same.”

(40) Y después hacerlo como legalmente (12B1).

“And then do it like legally.”

Regarding other syntactic positions of como in Spanish, there were two tokens embedded in noun phrases (41) and four tokens embedded in prepositional phrases (42).

(41) Era una rosa como con brillitos (3B2).

“It was a rose like with sparkles.”

(42) Pero más para como turistas en ciertas partes (5B2).

“But more for like tourists in certain parts.”

There were also only a few tokens of como in Spanish in the verbal domain with one token between the subject and a verb (43), and another example between an auxiliary and a verb (44).

(43) Ellos solo como vinieron a mi casa (15B2).

“They only like came to my house.”

(44) Es la única manera que puedes como entrar (1B2).

“It is the only way that you can like enter.”

The use of como que in Spanish as a discourse particle within a clause was very infrequent, but there were three examples that modified a determiner phrase (45) and one example each that modified a prepositional phrase (46) and an adjectival phrase (47). There was also one token that appeared between the subject and the verb (48).

(45) Además quedaba como que algo para la comida, para los libros (12B2).

“Additionally, there was like something left for the food, for the books.”

(46) Estaba pensando como que en literature (12B2).

“I was thinking like about literature.”

(47) Estoy como que bien perdida (12B2).

“I am like quite lost.”

(48) Entonces ya verdad el XY como que entendió (1B2).

“Then the truth, XY like understood.”

In Spanish, like was not used as a discourse particle.

Table 5 summarizes the syntactic positioning of the use of like in English and the Spanish variants as discourse particles. These syntactic positions were identified for like in English by Andersen (2001). The use of like as a discourse particle in English most frequently modified determiner phrases (39.24%; 330/840). Its second most frequent positioning was before adjectival phrases (14.74%; 124/840), followed by the verbal domain between the subject and the verb (13.67%; 115/840). The use of como as a discourse particle in Spanish also most frequently modified determiner phrases, with more than half of the examples in this position (54.88%; 45/82). Its second most frequent positioning was before prepositional phrases (18.29%; 15/82), which occurred relatively infrequently for like in English (4.64%; 39/840). The discourse particle como also modified adjectival phrases (13.41%; 11/82), which was its third most frequent position. Unlike for like in English, there were only two tokens of the use of como in Spanish as a discourse particle in the verbal domain. The use of como que in Spanish as a discourse particle was very rare with only six tokens, half of which modified a determiner phrase (50%; 3/6). Lastly, tokens of like in Spanish only appeared outside of the clause as discourse markers. These results mirror the overall distributions of como que and like in Spanish as discourse markers and discourse particles, which both are primarily used as discourse markers, differing from like in English and como in Spanish.

Table 5.

Syntactic positions of like in English and como, como que, and like in Spanish as discourse particles.

Syntactic position	English	Spanish
	like	como	como que	like
Modifying a determiner phrase	39.24% (330/840)	54.88% (45/82)	50% (3/6)	0% (0/0)
Modifying a prepositional phrase	4.64% (39/840)	18.29% (15/82)	16.67% (1/6)	0% (0/0)
Modifying an adjectival phrase	14.74% (124/840)	13.41% (11/82)	16.67% (1/6)	0% (0/0)
Modifying an adverbial phrase	1.07% (9/840)	4.88% (4/82)	0% (0/6)	0% (0/0)
Between the subject and the verb	13.67% (115/840)	1.22% (1/82)	16.67% (1/6)	0% (0/0)
Between a modal and the verb	4.64% (39/840)	0% (0/82)	0% (0/6)	0% (0/0)
Between an auxiliary and a gerund/participle	7.97% (67/840)	1.22% (1/82)	0% (0/6)	0% (0/0)
Between the verb + to + infinitive	3.21% (27/840)	0% (0/82)	0% (0/6)	0% (0/0)
Embedded in a noun phrase	2.50% (21/840)	2.44% (2/82)	0% (0/6)	0% (0/0)
Embedded in a prepositional phrase	7.13% (60/840)	4.88% (4/82)	0% (0/6)	0% (0/0)
Embedded in an adjectival phrase	0.48% (4/840)	0% (0/82)	0% (0/6)	0% (0/0)
Passive voice constructions	0.59% (5/840)	0% (0/82)	0% (0/6)	0% (0/0)

In summary, although the use of like in English and como in Spanish are both used as discourse particles with a similar distribution, the use of like in English as a discourse particle occurs in many more distinct syntactic positions. Specifically, the use of like in English as a discourse particle occurs in 12 syntactic positions, whereas the use of como as a discourse particle occurs in eight syntactic positions, with more than half appearing in the determiner phrase. This result highlights an important distinction between like in English and como in Spanish. The use of como que in Spanish as a discourse particle is infrequent and like in Spanish is not used as a discourse particle.

Discussion

Overall, this qualitative and quantitative analysis of like in English and como, como que, and like in Spanish sheds light on the limited interaction of linguistic repertoires at the discourse level in the speech of bilinguals in a language contact situation in which English is the majority language. It would be expected that discourse-pragmatic features such as the discourse markers and discourse particles in this study would be highly susceptible to borrowing in contact situations because they appear on the periphery of grammar and are detachable (e.g., Brody, 1987, 1995; Dajko & Carmichael, 2014). Several previous studies of discourse markers in bilingual communities in the USA have concluded that discourse markers in English as the majority language are used alongside or at the expense of equivalent discourse markers in minority languages (Aaron, 2004; Flores-Ferrán, 2014; Goss & Salmons, 2000; Salmons, 1990; Torres, 2002; Torres & Potowski, 2008). Lipski (2005) further argues that the insertion of English discourse markers in Spanish discourse may be automatic and even against the preference of Spanish-English bilinguals. However, contact with the majority language of English in Southern Arizona does not appear to radically influence the use of como, como que, and like in Spanish in the speech of these bilinguals.

The most prominent evidence of contact with English in the speech of these bilinguals is the use of like in Spanish. However, like in Spanish is significantly less frequent (only 16.88 tokens per 10,000 words) than like in English (360.38 tokens per 10,000 words). Moreover, like in Spanish is conditioned significantly differently to like in English, because it is only used as a discourse marker and has not been incorporated into Spanish syntax as a discourse particle. Furthermore, there is no evidence that like has replaced other discourse markers in Spanish because there is a clear preference for como among the bilinguals in this study. There is no evidence of convergence to lessen the cognitive load (Matras, 1998, 2000), fill a gap (Hill & Hill, 1986; Hlavac, 2006), or mark or strengthen discourse boundaries (Dajko & Carmichael, 2014; de Rooij, 2000; Maschler, 1994, 1997, 2009).

The use of como and como que in Spanish in the speech of these bilinguals could also be the result of contact with like in English as calques. However, the use of como and como que in Spanish is significantly less frequent than the use of like in English. This result mirrors the results of Jørgensen and Stenström (2009), who found that the use of como in Madrid, a non-contact variety of Spanish, was less frequent than the use of like in British English. It also reflects the results of Sankoff et al. (1997), who found that comme in French was used less frequently than like in English by the same French-English bilinguals. In addition to a difference in the overall frequencies of these variants in both languages, the use of como and como que in Spanish are conditioned differently than like in English in the speech of the same bilinguals. The distribution of como que in Spanish as a discourse marker and as a discourse particle is significantly different to like in English because it is used primarily as a discourse marker. Moreover, the use of como que as a discourse marker in Spanish only modifies matrix clauses and only appears on the left periphery, unlike the use of like as a discourse marker in English, which can modify both matrix clauses and subordinate clauses and can appear on the left or right peripheries. Although como in Spanish is conditioned similarly to like in English in its overall distribution as a discourse marker and as a discourse particle, the use of como in Spanish is not as widespread as the use of like in English. Mirroring como que, the use of como as a discourse marker only appears before matrix clauses and on the left periphery. The use of como as a discourse particle occurs in fewer syntactic positions than like in English, with over half of the tokens occurring in the determiner phrase.

The overall differences in the use of like in English and como, como que, and like in Spanish suggest that these four variants could be at different stages in their development in the speech of the bilinguals of the present study and highlight the importance of identifying the three Spanish variants equivalent to like in English and analyzing them separately. While like in English appears to be conditioned differently to the Spanish variants in terms of its distribution as a discourse marker or discourse particle and its syntactic positioning, the three Spanish variants also appear to be conditioned differently. The use of like in English occurs frequently as both a discourse marker and a discourse particle, and its use is widespread in terms of the syntactic positions in which it can appear. The use of como in Spanish occurs less frequently as both a discourse marker and discourse particle and in fewer syntactic positions. The use of como que is less frequent than that of como, corroborating Jimenez and Flores-Ferrán’s (2018) similar results for speakers of Mexican varieties. Distinct from like in English and como in Spanish, como que in Spanish is used almost exclusively as a discourse marker rather than as a discourse particle. Lastly, like in Spanish is very infrequent and is only used as a discourse marker. A hypothesis of the development of the three Spanish variants is presented visually in Figure 7, corresponding to D’Arcy’s (2005) proposal regarding the grammaticalization of like in English, first from a discourse marker to a discourse particle in the determiner phrase, then to a discourse particle in the verbal phrase (p. 210).

Figure 7.

Hypothesis of the development of like in English and como, como que, and like in Spanish among bilinguals.

An analysis of como, como que, and like in Spanish in apparent time is necessary to explore this hypothesis of their development. Future studies should also continue to track the use of these variants in US Spanish as discourse particles to assess if they extend to more syntactic positions. The positive correlation between the percent use of como and the combined Spanish variants as discourse particles and Spanish dominance also warrants further investigation. Spanish-dominant bilinguals appear to be the leaders in the use of como in Spanish as a discourse particle. This result may indicate their knowledge of Spanish syntax to be able to incorporate como within the sentence (echoing Poplack, 1980). Another possible explanation is that Spanish-dominant speakers use como less frequently as a discourse marker, which would elevate their percent use of como as a discourse particle when compared with English-dominant speakers, who use como more frequently as a discourse marker. For now, the results of this study on the use of like in English and como, como que, and like in Spanish provide a snapshot in a moment in time of their development.

Conclusion

This study has analyzed the use of like in English and como, como que, and like in Spanish in the speech of 18 Spanish-English bilinguals from Southern Arizona. Several previous studies have concluded that discourse markers are subject to borrowing in contact situations (e.g., Brody, 1987, 1995; Dajko & Carmichael, 2014). These studies include analyses of the use of English discourse markers alongside or at the expense of discourse markers in minority languages in bilingual communities across the USA (Aaron, 2004; Flores-Ferrán, 2014; Goss & Salmons, 2000; Lipski, 2005; Salmons, 1990; Torres, 2002; Torres & Potowski, 2008). The results of this study on the use of like in English and its three Spanish equivalents in the speech of bilinguals from Southern Arizona shed light on the limited interaction of linguistic repertoires in the speech of bilinguals, even in language contact situations in which one language is dominant. The use of like in English as a discourse marker and discourse particle is more frequent and conditioned differently than the use of como, como que, and like in Spanish. In addition, the use of como, como que, and like in Spanish each appear to be conditioned differently, contributing to our knowledge of the Spanish equivalents of like in English.

The Spanish heritage speakers in this study represent a special population of bilinguals who live in a world of two languages, often divided by personal and professional contexts. They reported that they most often use Spanish with family and friends, while they use English at school and work. A hallmark of the language use of these participants is a rich repertoire of discourse-pragmatic features in both English and Spanish, including the discourse markers and discourse particles analyzed in this study. This analysis of the use of like in English and como, como que, and like in Spanish in the speech of these bilinguals highlights their sophisticated linguistic competence and their knowledge of the subtleties of the use of these features in the two languages they speak.

Footnotes

Acknowledgements

I would like to begin by thanking the participants of the study. I would also like to acknowledge and thank the following people: Ana Carvalho, Perry Gilmore, Lillian Gorman, Norma Mendoza-Denton, Joseph Casillas, Tasha Campbell, Ryan Bessett, Justin Paz, Marta Ramírez Martínez, Yamile Díaz, Giovanni Nino, Ashley Scott, Hannah Bolados, Kristin Helland, Marcy Ziska, Kristy Doran, Brigette Walters, and Claire Lozano. All errors remain my own.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Author biography

Joseph Kern, PhD, is an Assistant Professor of Spanish at the University of Virginia’s College at Wise. His research explores discourse-pragmatic variation in Spanish, especially in varieties of Spanish in contact with English in the U.S. Southwest and with Portuguese on the border between Uruguay and Brazil. He has analyzed several discourse-pragmatic variables in Spanish including quotatives, discourse markers, general extenders, and tag questions.

References

Aaron

J. E.

(2004). “So respetamos un tradición del uno al otro.” Spanish in Context, 1, 161–179.

Andersen

(2001). Pragmatic markers and sociolinguistic variation: A relevance-theoretic approach to the language of adolescents. Amsterdam/Philadelphia: John Benjamins.

Andersen

(2014). Pragmatic borrowing. Journal of Pragmatics, 67, 17–33.

Birdsong

Gertken

L. M.

Amengual

(2012). Bilingual language profile: An easy-to-use instrument to assess bilingualism. COERLL, University of Texas at Austin. Retrieved from https://sites.la.utexas.edu/bilingual

Brinton

L. J.

(2006). Pathways in the development of pragmatic markers in English. In van Kemenade

Los

(Eds.), The handbook of the history of English (pp. 306–334). Oxford: Blackwell.

Brody

(1987). Particles borrowed from Spanish as discourse markers in Mayan languages. Anthropological Linguistics, 29, 507–521.

Brody

(1995). Lending the ‘unborrowable’: Spanish discourse markers in indigenous American languages. In Silva-Corvalán

(Ed.), Spanish in four continents: Studies in language contact and bilingualism (pp. 132–148). Washington, DC: Georgetown University Press.

Carvalho

A. M.

(2012–). Corpus del Español en el Sur de Arizona (CESA). University of Arizona. Retrieved from http://www.cesa.arizona.edu

D’Arcy

(2005). Like: Syntax and development (Unpublished doctoral dissertation). University of Toronto, Toronto, Canada.

10.

D’Arcy

(2017). Discourse-pragmatic variation in context: Eight hundred years of LIKE. Amsterdam/Philadelphia: John Benjamins.

11.

Dailey-O’Cain

(2000). The sociolinguistic distribution of and attitudes toward focuser like and quotative like. Journal of Sociolinguistics, 4, 60–80.

12.

Dajko

Carmichael

(2014). But qui c’est la différence? Discourse markers in Louisiana French: The case of but vs. mais. Language in Society, 43, 159–183.

13.

de Rooij

V. A.

(2000). French discourse markers in Shaba Swahili conversations. International Journal of Bilingualism, 4, 447–466.

14.

Erker

Bruso

(2017). Uh, bueno, em…: Filled pauses as a site of contact-induced change in Boston Spanish. Language Variation and Change, 29, 205–244.

15.

Fleischman

Yaguello

(2004). Discourse markers across languages? Evidence from English and French. In Moder

C. L.

Martinovic-Zic

(Eds.), Discourse across languages and cultures (pp. 129–147). Amsterdam: John Benjamins.

16.

Flores-Ferrán

(2014). So pues entonces: An examination of bilingual discourse markers in Spanish oral narratives of personal experience of New York City-born Puerto Ricans. Sociolinguistic Studies, 8, 57–83.

17.

Fuller

J. M.

(2003). Discourse marker use across speech contexts: A comparison of native and non-native speaker performance. Multilingua, 22, 185–208.

18.

Goss

E. L.

Salmons

J. C.

(2000). The evolution of a bilingual discourse marking system: Modal particles and English markers in German-American dialects. International Journal of Bilingualism, 4, 469–484.

19.

Hill

J. H.

Hill

K. C.

(1986). Speaking Mexicano: Dynamics of syncretic language in Central Mexico. Tucson: University of Arizona Press.

20.

Hlavac

(2006). Bilingual discourse markers: Evidence from Croatian-English code-switching. Journal of Pragmatics, 38, 1870–1900.

21.

Jaramillo

J. A.

(1995). The passive legitimization of Spanish: A macrosociolinguistic study of a quasi-border: Tucson, Arizona. International Journal of the Sociology of Language, 114, 67–91.

22.

Jimenez

Flores-Ferrán

(2018). The functions of the Spanish approximators como and como que in institutional and non-institutional discursive contexts. Sociocultural Pragmatics, 6, 145–171.

23.

Jørgensen

A. M.

Stenström

A. B.

(2009). Dos marcadores pragmáticos contrastados en el lenguaje juvenil: El inglés like y el español como. Español Actual, 92, 103–122.

24.

Kern

(2014). Como in commute: The travels of a discourse marker across languages. Studies in Hispanic and Lusophone Linguistics, 7, 279–298.

25.

Labov

(1972). Sociolinguistic patterns. Philadelphia: University of Pennsylvania Press.

26.

Lipski

(2005). Code-switching or borrowing? No sé so no puedo decir, you know. In Sayahi

Westmoreland

(Eds.), Selected proceedings of the Second Workshop on Spanish Sociolinguistics (pp. 1–15). Somerville, MA: Cascadilla Proceedings Project.

27.

Macaulay

(2002). Discourse variation. In Chambers

J. K.

Trudgill

Schilling-Estes

(Eds.), The handbook of language variation and change (pp. 283–305). Oxford: Blackwell.

28.

Maschler

(1994). Metalanguaging and discourse markers in bilingual conversation. Language in Society, 23, 325–366.

29.

Maschler

(1997). Emergent bilingual grammar: The case of contrast. Journal of Pragmatics, 28, 279–313.

30.

Maschler

(2009). Metalanguage in interaction: Hebrew discourse markers. Amsterdam/Philadelphia: John Benjamins.

31.

Mata

(2016). Traversing the wall: A study of language contact among heritage and immigrant speakers of Spanish in the Tijuana-San Diego border area (Doctoral dissertation). University of California, San Diego, San Diego, California.

32.

Matras

(1998). Utterance modifiers and universals of grammatical borrowing. Linguistics, 36, 281–332.

33.

Matras

(2000). Fusion and the cognitive basis for bilingual discourse markers. International Journal of Bilingualism, 4, 505–528.

34.

Miller

(2009). Like and other discourse markers. In Peters

Collins

Smith

(Eds.), Comparative studies in Australian and New Zealand English: Grammar and beyond (pp. 315–336). Amsterdam: John Benjamins.

35.

Miller

Weinert

(1995). The function of like in dialogue. Journal of Pragmatics, 23, 365–393.

36.

Muysken

(2000). Bilingual speech: A typology of code-mixing. Cambridge/New York: Cambridge University Press.

37.

Myers-Scotton

(1993). Duelling languages: Grammatical structure in codeswitching. Oxford/New York: Clarendon Press.

38.

Peterson

Vaattovaara

(2014). Kiitos and pliis: The relationship of native and borrowed politeness markers in Finnish. Journal of Politeness Research, 10, 247–269.

39.

Pichler

(2010). Methods in discourse variation analysis: Reflections on the way forward. Journal of Sociolinguistics, 14, 581–608.

40.

Pichler

(2013). The structure of discourse-pragmatic variation. Amsterdam/Philadelphia: John Benjamins.

41.

Poplack

(1980). Sometimes I’ll start a sentence in Spanish Y TERMINO EN ESPAÑOL: Toward a typology of code-switching. Linguistics, 18, 581–618.

42.

Poplack

Levey

(2009). Contact-induced grammatical change: A cautionary tale. In Auer

Schmidt

J. E.

(Eds.), Language and space: An international handbook of linguistic variation (pp. 391–419). Berlin: Mouton de Gruyter.

43.

Poplack

Sankoff

(1984). Borrowing: The synchrony of integration. Linguistics, 22, 99–135.

44.

R Core Team (2013). R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. Retrieved April 22, 2016 from http://www.R-project.org

45.

Real Academia Española (2015). Diccionario de la lengua española (23rd Edition). Madrid: Espasa. Retrieved April 22, 2016 from http://www.rae.es

46.

Rickford

J. R.

McNair-Knox

(1994). Addressee- and topic-influenced style shift: A quantitative sociolinguistic study. In Biber

Finegan

(Eds.), Sociolinguistic perspectives on register (pp. 235–276). New York: Oxford University Press.

47.

Romaine

Lange

(1991). The use of like as a marker of reported speech and thought: A case of grammaticalization in progress. American Speech, 66, 227–279.

48.

Said-Mohand

(2008). Aproximación sociolingüística al uso del marcador del discurso como en el habla de jóvenes bilingües en la Florida. Revista Internacional de Lingüística Iberoamericana, 6(2), 71–93.

49.

Salmons

(1990). Bilingual discourse marking: Code switching, borrowing, and convergence in some German-American dialects. Linguistics, 28, 453–480.

50.

Sánchez Muñoz

. (2007). Style variation in Spanish as a heritage language: A study of discourse particles in academic and non-academic registers. In Potowski

Cameron

(Eds.), Spanish in contact: Policy, social and linguistic inquiries (pp. 153–171). Amsterdam: John Benjamins.

51.

Sankoff

Thibault

Nagy

Blondeau

Fonollosa

M. O.

Gagnon

(1997). Variation in the use of discourse markers in a language contact situation. Language Variation and Change, 9, 191–217.

52.

Schiffrin

(1987). Discourse markers. Cambridge: Cambridge University Press.

53.

Schourup

L. C.

(1985). Common discourse particles in English conversation. New York: Garland.

54.

Tagliamonte

(2005). So who? Like how? Just what?: Discourse markers in the conversations of Young Canadians. Journal of Pragmatics, 37, 1896–1915.

55.

Tagliamonte

(2012). Variationist sociolinguistics: Change, observation, interpretation. Malden: Wiley-Blackwell.

56.

Thomason

S. G.

Kaufman

(1988). Language contact, creolization and genetic linguistics. Berkeley: University of California Press.

57.

Torres

(2002). Bilingual discourse markers in Puerto Rican Spanish. Language in Society, 31, 65–83.

58.

Torres

Potowski

(2008). A comparative study of bilingual discourse markers in Chicago Mexican, Puerto Rican, and MexiRican Spanish. International Journal of Bilingualism, 12, 263–279.

59.

Traugott

E. C.

(1995). The role of the development of discourse markers in a theory of grammaticalization. Paper presented at the 12th International Conference on Historical Linguistics, Manchester.

60.

Treffers-Daller

(1994). Mixing two languages: French-Dutch contact in a comparative perspective. New York: Mouton de Gruyter.

61.

Underhill

(1988). Like is, like, focus. American Speech, 63, 234–246.

62.

Valdés

(2000). Bilingualism and language use among Mexican Americans. In McKay

S. L.

Wong

S. C.

(Eds.), New immigrants in the United States (pp. 99–137). Cambridge: Cambridge University Press.

63.

Van Hout

Muysken

(1994). Modeling lexical borrowability. Language Variation and Change, 6(1), 39–62.

64.

Zavala

(2001). Borrowing evidential functions from Quechua: The role of pues as a discourse marker in Andean Spanish. Journal of Pragmatics, 33, 999–1023.