Intensification Strategies in English–Spanish Bilingual Speech: Examining Lexical and Morphological Markers in Miami Bilinguals’ Discourse

Abstract

This paper investigates the speech variety of the English–Spanish community of Miami, which features a high degree of English–Spanish bilingualism. Specifically, it explores intensification, a site of analytic–synthetic conflict between English and Spanish grammars. English predominantly uses lexical-analytic strategies (e.g., very beautiful, a big house) for intensification, whereas Spanish employs more morphological-synthetic markers (e.g., guapísimo ‘very beautiful’, un casón ‘a big house’). Concretely, the current study aims to investigate whether Miami bilinguals have preferences in terms of the language or strategy of choice to express intensification and whether these preferences are influenced by intralinguistic (e.g., semantic-pragmatic function of the intensifier) or extralinguistic factors (e.g., speakers’ proficiency in, acquisition of, and attitudes toward both languages). To this end, an empirical study is conducted on three corpora, one bilingual and two monolingual ones. In this study, a wide variety of both analytic and synthetic intensifiers is found. The qualitative and quantitative findings reveal that Miami bilinguals use more English than Spanish intensifiers and favor analytic intensifiers over synthetic ones. However, among the Spanish intensifiers, the proportion of synthetic forms is significantly higher than among the English intensifiers.

1. Introduction

In an ever-multilingual world, attention within research is shifting more from the investigation of unilingual language varieties toward that of situations of multilingualism and language contact. Depending on various factors, contact situations can result in a variety of linguistic outcomes (Muysken, 2013), such as borrowing (Durkin, 2020), convergence (Heine & Kuteva, 2005), codeswitching (Muysken, 2000), and the formation of mixed languages (Bakker & Papen, 1997) and creoles (Hymes, 1971). Given these diverse possible outcomes of language contact, it is crucial to investigate a wide array of contact situations and multilingual communities to better understand the conditions shaping these phenomena. The current paper investigates the effects of language contact in an English–Spanish bilingual community as concerns the employed strategies of intensification (e.g., very beautiful, guapísimo).

An important notion to investigate possible outcomes of contact are conflict sites, which are grammatical structures and phenomena with respect to which the grammars of languages in contact differ substantially, prompting the necessity to find resolutions to these conflicts. Previous research identifies and investigates a variety of these English–Spanish conflict sites, such as subject pronoun expression (González-Vilbazo & Koronkiewicz, 2016), word order within both verb (Chan, 2008; Lakshmanan et al., 2016) and noun phrases (Parafita Couto & Gullberg, 2019; Balam & Parafita Couto, 2019; Stadthagen-González et al., 2019) and gender agreement within the noun phrase (Balam, 2016; Valdés Kroff, 2016; Królikowska et al., 2019; Bellamy & Parafita Couto, 2022). Another prominent, but understudied conflict site in English–Spanish bilingual speech concerns the fact that Spanish is a fusional synthetic language (Martínez-García et al., 2021), whereas English is more of an analytic language (Haspelmath & Michaelis, 2017). One relevant construction in this respect that has been studied before in English–Spanish bilingual speech is the diminutive construction (Vanhaverbeke & Enghels, 2021). Intensification, which is the focus of this paper, constitutes another such conflict site.

Intensification is often considered a linguistic universal (Wierzbicka, 1996; Rainer, 2015). One prototypical and ubiquitous instantiation of intensification concerns the scaling of adjectives and adverbs by means of degree adverbs (e.g., very beautiful, very quickly; Labov, 1984). Depending on one’s scope, however, intensification may encompass the scaling of virtually any linguistic element by a wide array of morphosyntactic intensifiers. This scaling can have several different semantic-pragmatic functions. On the one hand, it can amplify the intensity of the dimension, amount or degree expressed by an element, which are all considered quantitative functions (e.g., a megahouse). On the other hand, however, intensifiers can also express more qualitative-evaluative notions, both positive and negative (e.g., a superday, this fucking thing), and interpersonal or illocutionary connotations (e.g., for fuck’s sake; S. Schneider, 2017; Roels, 2024).

The community under consideration here is in the city of Miami, in southern Florida. This city is characterized by high proportions of Hispanic people in its population, many of them being Cubans. However, Miami’s Hispanic population is generally very diverse, including migrants from all over Latin America, including the Dominican Republic, Guatemala, Honduras, Nicaragua, Costa Rica, Panama, Venezuela, Colombia, Ecuador, and Argentina. It is because of this diversity of speakers and Miami’s key location at the crossroads of North America, Central America, and the Caribbean, in addition to the availability of ample high-quality corpus data, that Miami can be considered a very adequate setting to investigate the effects of various processes of language contact. As a consequence of its diverse Hispanic population, the city features high degrees of English–Spanish bilingualism (Oller & Eilers, 2002; D. Carter et al., 2011), including contact effects that are also at the center of this study, namely the broader concept of convergence (Lynch, 1999; P. M. Carter et al., 2020) and the more concrete phenomenon of codeswitching (Valdés Kroff, 2016; Blokzijl et al., 2017; Vanhaverbeke & Enghels, 2021).

To begin with, linguistic convergence encompasses different types of changes observed in languages in contact because of which said languages become structurally more similar (Birnbaum, 1985; Matras, 2011; Bouzouita & Enghels, 2021). Linguistic convergence can affect languages on different linguistic levels, including the phonological, morphological, and syntactic ones1. It is a form of indirect change, which has the ability to influence the structure of a language at a more abstract and profound level. In contact situations where convergence takes place, one of the languages contains a construction that is somewhat analogous to a construction in the other language, prompting bilingual speakers to equate them and, as such, causing them to become more similar to each other in terms of meaning, use, and frequency (Grant, 2020). This conflation of constructions from different languages is in line with the notion of diaconstructions from Diasystematic Construction Grammar (Höder, 2018), which postulates that bilingual speakers create abstract, language-unspecific constructions in their mental lexicon that encompass analogous structures from both languages. Interestingly, when several constructions alternate in one of the languages and the formation of such a diaconstruction increases the frequency of one of them, the balance between these constructions can shift and the frequency of competing constructions can decrease (Heine & Kuteva, 2005). This notion of convergence through the process of equating analogous structures in the languages in contact is highly relevant to the current study, because English and Spanish, as is explained throughout this paper, partly employ the same strategies and, in some cases, even the same specific items to intensify. This could prompt convergence effects and cause these shared strategies and specific intensifiers to become more frequent.

A second contact-induced phenomenon highly prominent in the speech of the bilingual community of Miami is codeswitching, defined as the alternating use of two or more languages within one conversation or utterance (Muysken, 2000). A distinction can be made between intersentential (1a), intrasentential (1b), and the more marginally attested intraword codeswitching (1c; Deuchar, 2006; Stefanich et al., 2019).

(1)	a.	I’m fine.	¿Tú	cómo	estás?
		I’m fine	you	how	be.2sg
		‘I’m fine. How are you?’ (Ghirardini, 2006, p. 17)

	b.	This	morning mi	hermano	y	yo	fuimos	a
		this	morning my	brother	and	I	go.1pl.pst	to
		comprar	some milk.
		buy	some milk
		‘This morning my brother and I went to buy some milk.’ (Belazi et al., 1994, p. 222)

	c.	Voy	a	hang-ear	con	mis	amigo-s.
		go.1sg	to	hang-vbz	with	my	friend-pl
		‘I’m going to hang with my friends.’ (Stefanich et al., 2019, p. 3)

There is now a consensus that codeswitching does not occur arbitrarily but is a rule-governed phenomenon. Over the years, researchers have tested numerous theories and constraints, some of them newly formulated and some of them already existent theories of language, to explain and predict which switches are grammatical and which are not (Poplack, 1980; Mahootian, 1993; Myers-Scotton, 1993; Bhatt & Bolonyai, 2011; Goldrick et al., 2016). Like convergence, this phenomenon is very relevant when investigating intensifying strategies in English–Spanish bilingual speech, as it results as one of the possible outcomes of a conflict site.

Against this background, it is of particular interest to investigate how intensification manifests under the influence of contact effects. First and foremost, as speakers have the intensifying systems of two languages at their disposal, which both use different strategies to intensify, this prompts some questions as to in which language and with which strategy (i.e., morphological-synthetic or lexical-analytic) bilingual speakers choose to express intensification. These choices are likely not random but can further be influenced by a variety of factors. The choice for one language over the other, for instance, can be influenced by the semantic-pragmatic function of the intensifier. In English–Spanish bilingual speech, English has indeed been shown to be preferred to express quantitative notions, whereas Spanish would be more reserved for qualitative-evaluative notions (Vinagre Laranjeira, 2005; Betti & Enghels, 2018). This is because, for most bilinguals, the language of their country of origin (which tends to be Spanish for English–Spanish bilinguals) is the language of subjectification, while the language of their country of residence (which tends to be English for English–Spanish bilinguals) is the language of objectification (Ohlson, 2007). This dichotomy strongly correlates with the situations in which speakers tend to use both languages (Blom & Gumperz, 1972), with the language of subjectification usually being spoken in the home and used to talk about the speaker’s culture, emotions, and interpersonal relations, and the language of objectification being spoken at school or work and used to discuss topics related to analytical thinking, problem-solving, and collaboration. The language of choice can also be influenced by certain constraints and theories of codeswitching that are postulated. Relevant examples of these are the Free Morpheme Constraint (Poplack, 1980), which predicts that, in bilingual discourse, bound morphemes of one language tend not to attach to stems in another language. Moreover, the Uniform Structure Principle (Myers-Scotton, 2002) predicts that certain grammatical structures align with the grammar of the matrix language (ML), which is often defined as the language of the finite verb (Myers-Scotton, 1993). An example of this can be observed in Jordanian Arabic–English codeswitching, where Arabic determiners are found with both English and Arabic nouns when Arabic is the ML (Myers-Scotton, 2002). Furthermore, the choice of one language over another is likely influenced by the sociolinguistic profile of the speaker (D. Carter et al., 2011; Parafita Couto et al., 2014), specifically by their language proficiency, age of acquisition, and attitudes. The first research question, as such, is which intralinguistic and extralinguistic factors influence the configuration of Miami bilinguals’ intensifying constructions. In concrete, the impact is investigated of the semantic-pragmatic function of the intensifier, relevant structural constraints and theories, as well as speakers’ self-reported proficiency in, age of acquisition of, and attitudes toward English and Spanish. These questions and relevant hypotheses are discussed more elaborately in Section 2.2.2.

Secondly, it is also of interest to investigate to what extent the intensifying systems of these two languages, which employ different strategies with different frequencies, display convergence effects, where the favored intensifying strategy or the frequency of certain strategies within these systems changes. To measure the possible convergence effects, this paper investigates to what extent the intensifying strategies used in the bilingual speech variety of Miami are different from those used in monolingual English-speaking and Spanish-speaking communities. The second research question, then, is to what extent Miami bilinguals’ intensification displays convergence effects as to the employed strategies and their frequencies when compared to monolingual English and Spanish varieties. Expected outcomes are higher frequencies of intensifying strategies or specific intensifying items that are shared between the two languages. In the case of English and Spanish, as further exemplified below, the shared strategy is analytic intensification. In order to test for convergence effects, conversational corpus data from Miami and from two monolingual corpora, namely the Santa Barbara Corpus and the Havana subcorpus of Ameresco, are analyzed qualitatively and quantitatively. This is elaborated upon in Section 2.2.3.

The remainder of the paper is structured as follows. Section 2 discusses the intensifying strategies used in English, Spanish, and bilingual speech. Next, Section 3 explains the methods used to answer the research questions, including the corpora consulted. Then, Section 4 contains the various results of the corpus study and the subsequent analyses and statistical tests performed. Within this section, Section 4.1 gives a descriptive overview of the results, Section 4.2 deals with the results concerning the influence of intralinguistic and extralinguistic variables and, finally, Section 4.3 addresses the results concerning convergence. To conclude, Section 5 relates how these results offer answers to the research questions.

2. Intensification Strategies

Intensification can be defined in many ways to encompass different types of morphophonological, morphosyntactic, pragmatic, and suprasegmental processes observed in discourse (Abdullaeva, 2024). The current study restricts its definition to those overt morphosyntactic elements that can be said to assign an intensifying or augmenting value to another overtly expressed element within the discourse (referred to as ‘bases’ throughout this paper), which both can belong to different word classes (e.g., a big house, me gusta mucho ‘I like it a lot’)2. Below, the ways in which these overt patterns of intensification are expressed are discussed, first in monolingual varieties of both English and Spanish (Section 2.1) and, second, in an English–Spanish bilingual speech variety (Section 2.2).

2.1. Intensification Strategies in English and Spanish Unilingual Modes

2.1.1. Intensification in English

In English, two word classes frequently intensified are adjectives (2a) and adverbs (2b). This is commonly achieved through the use of intensifying adverbs, of which very is one of the most prototypical ones. This class of lexical intensifiers, however, is highly susceptible to expansion, in English (Núñez Pertejo & Palacios Martínez, 2014; Aijmer, 2018; Schweinberger, 2021) and in other languages (ten Buuren et al., 2018; Roels & Enghels, 2020), through the addition of neologisms from a wide variety of sources, such as taboo words (2c). Other intensifiers for adjectives and adverbs include lexicalized multiword constructions, which can originate in nominal elements (2d) or prepositional phrases (2e; Emmer, 2020).

(2)	a.	She is very talented. She’s very smart. And she’ll be a force to be reckoned with. (COCA, SPOK, 2019)3
	b.	Happily, she did very well for us in the ER and was admitted to observation. (COCA, BLOG, 2012)
	c.	Here’s the thing about Justin Sherrell: He’s a fucking great drummer. (COCA, BLOG, 2012)
	d.	Because a thousand dollars here in New York City or San Francisco is a lot different than anywhere else. (COCA, SPOK, 2019)
	e.	Obstruction of justice in this situation is absurd. I fought hard as hell to make sure Mueller could do his job. (COCA, SPOK, 2019)

Other grammatical elements in English that can be intensified, are nouns (3a), verbs, or verb phrases (3b) and, to a more restricted degree, prepositional phrases (3c) and subordinate clauses (3d). In examples (3b–d), the intensifier has a quantitative function and amplifies the degree of the quality expressed by the base. For nominal bases, however, an intensifier in English generally expresses a qualitative-evaluative notion, as in (3a), in which the intensifier provides a negative evaluation.

(3)	a.	This is why I don’t like you. You fucking man whore. (COCA, WEB, 2012)
	b.	In the abstract, Americans really like the idea of rich Americans paying more in taxes. (COCA, SPOK, 2019)
	c.	So I’m, I’m always very, very against any sort of squelching of free speech. (COCA, SPOK, 2017)
	d.	And at the same time, it kind of feels very like it’s short-circuiting your brain sometimes. (COCA, SPOK, 2015)

Like its diminutive system (K. P. Schneider, 2003; Vanhaverbeke & Enghels, 2021), the intensification system of English relies for the largest part on lexical items. However, it also includes a limited paradigm of intensifying prefixes, such as super-, hyper-, mega-, and ultra-, of which super- is the most common (Palacios Martínez & Núñez Pertejo, 2014). These prefixes are used to intensify different types of bases, including adjectives (4a), where they express a high degree, and nouns (4b), where they express a large size or qualitative, generally positive, evaluation. Note that the status of these forms as true prefixes is disputed in many languages, both English (Palacios Martínez & Núñez Pertejo, 2012; Chválová, 2021) and others (Foltran & Nóbrega, 2016; Napoli & Ravetto, 2017), with some authors considering them adjectives (when intensifying nouns) and adverbs (when intensifying adjectives or adverbs). In the current study and based on etymological grounds, they are considered prefixes. This further allows for their analysis in contrast to true intensifying adverbs, such as very and really.

(4)	a.	I think the digs record was supercool because Kristen Webb was so good. (COCA, NEWS, 2016)
	b.	One of you has a superfan so super that they have a tattoo devoted to you. (COCA, SPOK, 2014)

As opposed to this paradigm of prefixes, certain varieties of colloquial English also feature two marginal, paradigmatically unrelated intensifying suffixes. The first of these is the suffix -ass. Originating in African American Vernacular English, the suffix has since been recognized as an established form in General American English (Cannon, 1979). It can be used on adjectives to quantitatively intensify the degree they express (5a). The suffix can also nominalize adjectives, in which case it describes a person who possesses or is characterized by the quality expressed by the adjective (5b). Especially in this latter sense, the suffix tends to convey a negative evaluation. The second suffix to consider is the marginally occurring form -zilla, generally restricted to more informal and slang registers (Morshed, 2018; Zahradníková, 2018). The suffix is used on nouns and has a negative qualitative-evaluative meaning (5c).

(5)	a.	I would’ve never worn those ugly-ass shoes. (COCA, SPOK, 2015)
	b.	Unfortunately, being a dumbass does not preclude you from reproducing. (COCA, BLOG, 2012)
	c.	I know you call me a bridezilla, Nicky, but I am not that crazy. (COCA, SPOK, 2010)

2.1.2. Intensification in Spanish

Spanish also uses analytic lexical forms to intensify a range of elements, but unlike English, Spanish employs synthetic intensification, especially in the form of suffixes, in a more productive way. First, various elements can function as analytic intensifiers in Spanish, such as adverbs (in the examples used with an adjective (6a) and a verb (6b) as bases), nominal elements (6c; here used on a verb), prepositional phrases (6d; here used on an adjective), or adjectives (6e; here used on a noun). In all but the last example, the intensifier quantitatively amplifies the degree expressed by the adjective or verb, while in (6e), the intensifier offers a qualitative evaluation, here negative.

(6)	a.	Ahí	donde	yo	vivo	es	muy	tranquilo	pero	a
		there	where	I	live.1sg	be.3sg	very	calm	but	at
		la	vez	es	muy	aburrido.
		the	time	be.3sg	very	boring
		‘There where I live it’s very calm, but at the same time, it’s very boring.’ (PRESEEA, LHAB_H11_001)4

	b.	A	mí	no	me	gusta	mucho	el	verano,	por	eso
		to	me	not	me	please.3sg	much	the	summer	for	that
		me	alegro	cuando		empieza	el	invierno.
		me	be.happy.1sg	when		begin.3sg	the	winter
		‘I don’t like summer very much, therefore I’m happy when winter begins.’ (PRESEEA, LHAB_M13_079)

	c.	De	hecho	cuando	falleció	sí	me	dolió	un	montón.
		in	fact	when	die.3sg.pst	yes	me	hurt.3sg.pst	a	heap
		‘In fact, when she died, it hurt me a lot.’ (PRESEEA, MEXI_H22_054)

	d.	Me	decidí	e	hice	una	cas-ita	allí	en,
		me	decide.1sg.pst	and	make.1sg.pst	a	house-dim	there	in
		en	Guïnes,	una	cas-ita	de	lo	más	bonita.
		in	Guïnes	a	house-dim	of	the	most	beautiful
		‘I decided to build a little house there in Güines, a very beautiful little house.’ (PRESEEA, LHAB_H32_061)

	e.	Les	voy	a	dar	a	mis	hijo-s	todo	lo	que
		them	go.1sg	to	give	to	my	child-pl	all	that	which
		su	puto	padre	no	nos	dio,	el	culero.
		their	fucking	father	not	us	give.3sg.pst	the	asshole
		‘I’m going to give my children everything their fucking father didn’t give us, the asshole.’ (PRESEEA, PUEB_M21_056)

Next, the Spanish paradigm of intensifying prefixes is similar to that of English, with prefixes like super-, hiper-, mega- and ultra-, of which super- is again the most frequent. Overall, however, they are all used more frequently than in English (Palacios Martínez & Núñez Pertejo, 2014). The prefixes can be used on various types of bases, such as adjectives (7a) and nouns (7b). The paradigm also contains forms not found in English, namely the prefixes re-, rete-, requete- and recontra- (7c; Martín García, 1998; Pharies, 2009).

(7)	a.	No	es	caro,	o	sea	está	bien	de	precio,
		not	be.3sg	expensive	or	be.3sg.sjv	be.3sg	well	of	price
		a	precio-s	regulares,	o	sea	tampoco	es	mega-barato.
		at	price-pl	regular	or	be.3sg.sjv	nor	be.3sg	ints-cheap
		‘It’s not expensive, I mean it’s well-priced, at regular prices, I mean it’s also not megacheap.’ (PRESEEA, MXLI_M13_029)

b.

El

sábado

y

el

domingo

en

un

motel

o

en

el

the

Saturday

and

the

Sunday

in

a

motel

or

in

the

único

motel

que

hay

en esa

parte

de

Pinar del Río

only

motel

that

be.3sg

in that

part

of

Pinar del Río

ponían

un

poqu-ito

de

música

y

todos

los muchacho-s

put.3pl.impf

a

little-dim

of

music

and

all

the boy-pl

bajaban

y

esa

era

la

super-discoteca

de

ahí.

descend.3pl.impf

and

that

be.3sg.impf

the

ints-discotheque

of

there

‘On Saturdays and Sundays in a motel or in the only motel there was around that part of Pinar del Río, they put a little music and all the boys went down and that was the superdisco around there.’ (PRESEEA, LHAB_M11_007)

c.

Pues

yo

creo

que

fue

re-bién

para

mí.

Para

mi

so

I

believe.1sg

that

be.3sg.pst

ints-well

for

me

for

my

vida

me

ha

ayudado

mucho mucho

mucho

mucho.

Yo

life

me

have.3sg

helped

much much

much

much

I

creo

que

ha

sido

de

las

época-s

más

padres.

believe.1sg

that

have.3sg

been

of

the

time-pl

most

cool

‘So I think it was very good for me. For my life, it’s helped me very, very, very much. I think it’s been one of the coolest times.’ (PRESEEA, MEXI_H22_054)

As was stated before, Spanish also has a wide variety of intensifying suffixes. First, the suffix -ísimo is mainly used with adjectives and adverbs as bases and expresses a high degree of said adjective (8a). However, it can marginally be used on other types of bases as well, such as nouns (8b). The part -is- of the suffix can be reduplicated to express an even greater degree (8c).

(8)

a.

¿Momento-s

importantes

de

la

vida?

Eeh

los

dos

nacimiento-s

moment-pl

important

of

the

life

uh

the

two

birth-pl

de

mis

hijo-s,

esos

son

important-ísimos,

esos

son

of

my

child-pl

those

be.3sg

important-sup

those

be.3sg

el

momento

más,

los dos momento-s

más

importantes de mi

vida.

the

moment

most

the two moment-pl

most

important of my

life

‘Important moments in life? Uh the two births of my children, those are very important, those are the two most important moments of my life.’ (PRESEEA, LHAB_H21_013)

b.

Entonces

como

yo,

somos

siete

hermano-s

nunca

tuve

so

since

I

be.1pl

seven

sibling-pl

never

have.1sg.pst

la

gran

necesidad

de

buscar

amigo-s,

creo

que

ninguno

the

big

need

of

find

friend-pl

believe.1sg

that

none

de

mis

hermano-s

porque

los,

de

los

mismos

siete

yo

of

my

sibling-pl

because

the

of

the

same

seven

I

no

conozco

que

alguno

tenga

un

gran

amigu-ísimo.

not

know.1sg

that

any

have.3sg.sjv

a

big

friend-sup

‘So since we’re seven siblings I never felt a great need to look for friends, I think none of my siblings because I don’t think any of the seven has a very great friend.’ (PRESEEA, GUAD_H22_002)

	c.	De	repente	hace	much-is~is~ísima			calor	y	luego	se
		of	sudden	do.3sg	much-red~red~sup			heat	and	later	itself
		va	el	calor	y	aaay	un	fri-azo.
		go.3sg	the	heat	and	ay	a	coldness-aug
		‘Suddenly there’s very, very, very much heat and later the heat goes away and ay (there’s) a great, intense coldness.’ (PRESEEA, GUAD_M21_062)

Next, Spanish has a rich paradigm of augmentative suffixes. These include the suffixes -ón, -azo, -ote, -udo, -aco, -acho, -astro, and -asco, of which the first three are the most frequent. As opposed to the superlative, which takes adjectives as its most prototypical base, these suffixes are primarily used on nouns; they can attribute a large size (9a) or a qualitative evaluation, either positive (9b) or negative (9c), to a base. Various of these suffixes can also be used on different types of bases, such as adjectives (9d) and nominalized verb forms (9e). Similar to the superlative, augmentative suffixes can be stacked to further emphasize the magnitude (9f).

(9)	a.	Pues	llevamos	un	canast-ón	y	ahí	los	echamos	todos.
		so	take.1pl	a	basket-aug	and	there	them	throw.1pl	all
		‘Then we take a big basket and we throw them all in there’ (PRESEEA, GUAD_H11_086)

	b.	Aparte	que	era	guap-ísimo,	un	cuerp-azo	y	el	señor
		except	that	be.3sg.impf	beautiful-sup	a	body-aug	and	the	mister
		bajaba	en	mallon-es	verdes	o	azules	de	aquí.
		descend.3sg.impf	in	pantyhose-pl	green	or	blue	from	here
		‘Except he was very handsome, a great, attractive body, and the man would come down from here in green or blue tights.’ (PRESEEA, MEXI_M23_024)

	c.	Pues	estaba	nacid-ito	mi	niño,	yo	lo	metí		al
		so	be.3sg.impf	born-dim	my	child	I	him	put.1sg.pst		in.the
		carro	para	que	no	fuera	tanto	ruid-azo		ahí.
		car	for	that	not	be.3sg.impf.sjv	so.much	noise-aug		there
		‘So my child was newborn, I put him in the car so that there wasn’t so much loud, unpleasant noise there.’ (PRESEEA, GUAD_M31_034)

	d.	Había	unos	arbol-ote-s	enormes	grand-otes	y	ahí
		be.3sg.impf	some	tree-aug-pl	enormous	big-aug	and	there
		había	un	montón	de	zopilote-s.
		be.3sg.impf	a	heap	of	vulture-pl
		‘There were some very huge, enormous trees and there there were a lot of vultures.’ (PRESEEA, MEXI_M31_108)

	e.	Físicamente	yo	las	encuentro	maravillosas,	espiritualmente	son	un
		physically	I	them	find.1sg	marvelous	spiritually	be.3pl	a
		poco	pele-onas	y	eso.
		little	fight-aug	and	that
		‘Physically I find them marvelous, spiritually they’re a little quarrelsome and all that.’ (PRESEEA, LHAB_H22_049)

	f.	Entonces	tiene	un	cas-on-on-on-ón			y	ya	están
		so	have.3sg	a	house-aug-aug-aug-aug			and	already	be.3pl
		los	dos	viej-illo-s	solos	con	los	sirviente-s.
		the	two	old-dim-pl	alone	with	the	servant-pl
		‘So he has a very, very, very big house and the two little old ones are there alone with the servants.’ (PRESEEA, MONR_M32_094)

2.1.3. Similarities and Differences Between Intensification in English and Spanish

As can be seen from the above descriptions, the intensifying paradigms of English and Spanish display some similarities and differences. Both languages can intensify a variety of different word classes. The forms used to intensify these elements can be lexical items of various types, but both languages can also employ affixes to this end. It is in these affix systems that the most important differences between both languages are observed. Although both affix systems can be observed to contain similar forms, including prefixes such as super- and mega-, the Spanish morphological system surpasses the English both in token and type frequency. It is generally observed that, even when considering these shared prefixes, synthetic forms have a higher frequency in Spanish than they do in English (Núñez Pertejo & Palacios Martínez, 2014). Furthermore, among its intensifying prefixes, Spanish has some forms not found in English, such as re-. Finally, in its suffixes, Spanish also has an elaborate paradigm, which is overall lacking in English.

This greater productivity of synthetic forms in Spanish compared to the prevalence of lexical and hence analytic intensifiers in English constitutes a potential conflict site in English–Spanish bilingual speech. Additionally, the varying frequencies of the analytic and synthetic strategies between the two languages can lead to convergence effects in this particular contact situation. The possibilities to express intensification in a bilingual setting are explored in more detail in the following subsection.

2.2. Intensification in Bilingual Modes

2.2.1. Previous Accounts

The unilingual paradigms of intensification of English and Spanish make for many ways in which different intensifying strategies can be combined in a bilingual setting. So far, the realization of intensification in a bilingual context is hardly touched upon in the literature. Fiorentini and Sansó (2017) investigate intensifiers in a Ladin–Italian bilingual setting in Trentino-South Tyrol in Italy, but focus more on discourse markers that intensify the speaker’s epistemic stance or the illocutionary force of the discourse, such as Ladin de segur ‘definitely’. Through the analysis of various types of intensifiers in terms of their borrowability, the study finds that these types of intersubjective intensifiers are used in bilingual speech the most easily. The current study, however, instead focuses on the intensification of explicit linguistic constituents within the discourse, rather than of the epistemic stance toward the discourse in a more general sense.

Additionally, Fernández Flórez (2022) analyzes the intensifiers muy and bien in the English–Spanish bilingual community of Tucson in Arizona, but ultimately concentrates on the use of these intensifiers in unilingual Spanish speech. It is found that these two intensifiers display similar behavior as they do in monolingual Spanish communities, both in their frequencies and the contexts in which they are used. Although these two specific lexical items are instances of the types of intensifiers that are within the scope of the current paper, the focus here is broadened to include all different types of lexical and morphological intensifiers, as well as contexts featuring English, Spanish, or a mix of the two.

In this way, the current paper aims to create a more comprehensive image of morphosyntactic intensification across its full breadth and across different contexts of language usage, both unilingual and bilingual, within a bilingual community. Accordingly, the study addresses two central research questions, as outlined in Section 1. Specifically, the first research question examines the influence of various intralinguistic and extralinguistic factors on the configuration of intensifying constructions among Miami bilinguals. The second question explores the extent to which Miami bilinguals’ intensifying strategies exhibit convergence effects.

2.2.2. Hypotheses Regarding Influence of Intralinguistic and Extralinguistic Factors on Intensification

First of all, it is of interest whether bilingual speakers have preferences in terms of the language in which or the strategy with which they express their intensifying constructions. Speakers have both English and Spanish intensifiers at their disposal and can use these on either English or Spanish bases. This leads to two possibilities for unilingual intensifying constructions, namely unilingual English and Spanish constructions, and two possibilities for intensifying constructions with a codeswitch in them, namely one with an English base modified by a Spanish intensifier and vice versa. These four possible configurations are shown in

Table 1. Possible combinations of languages and strategies of intensifiers.

	English Base	Spanish Base
English intensifier	very cheap	very barato
Spanish intensifier	muy cheap	muy barato

.

The number of possible configurations is furthermore expanded by taking into account the intensifying strategy and the matrix language (ML) of the clause containing the intensifying construction. The inclusion of these factors causes the number of configurations to increase to 16 possible constructions, which are shown in

Table 2. Possible combinations of ML, base language, and intensifier language/strategy.

		English ML		Spanish ML
		Analytic	Synthetic	Analytic	Synthetic
English base	English intensifier	The book is very cheap.	The book is hypercheap.	El libro es very cheap.	El libro es hypercheap.
	Spanish intensifier	The book is muy cheap.	The book is cheapísimo.	El libro es muy cheap.	El libro es cheapísimo.
Spanish base	English intensifier	The book is very barato.	The book is hyperbarato.	El libro es very barato.	El libro es hyperbarato.
	Spanish intensifier	The book is muy barato.	The book is baratísimo.	El libro es muy barato.	El libro es baratísimo.

.

The current study investigates the distribution of these patterns, including the prevalence of some of these over others, by considering the influence of various factors. A first parameter is the semantic-pragmatic function (i.e., purely quantitative or qualitative-evaluative) of the intensifier. Specifically in an English–Spanish bilingual context like Miami, Spanish is often used to express more affective, subjective notions, whereas English is used for more objective statements (Vinagre Laranjeira, 2005), which is a type of metaphorical codeswitching (Blom & Gumperz, 1972). This prompts the expectation that, if this principle is at play in the bilingual production of an intensifying construction, a quantitative intensifier is more likely to be in English, while a qualitative-evaluative one would rather be in Spanish. Some hypothetical examples of this can be seen in (10a–b), which show two constructions involving English quantitative intensifiers, and (11a–b), which show two constructions involving Spanish qualitative-evaluative intensifiers.

(10)	a.	Corrí	a	la	escuela	very	quickly.
		run.1sg.pst	to	the	school	very	quickly
		‘I ran to school very quickly.’
	b.	Esa	es	una	very	big	casa.
		that	be.3sg	a	very	big	house
		‘That is a very big house.’

(11)	a.	I’m tired because of the puto			day	that I’ve had.
		I’m tired because of the fucking			day	that I’ve had
		‘I’m tired because of the fucking day that I’ve had.’
	b.	Tengo	un	plan-azo	for my birthday tomorrow.
		have.1sg	a	idea-aug	for my birthday tomorrow
		‘I have a great idea for my birthday tomorrow.’

The choices that bilingual speakers make in this respect can also be conditioned by structural constraints and theories, such as the Free Morpheme Constraint and the Uniform Structure Principle5. It is expected that, if the Free Morpheme Constraint holds true for the data, intensifying constructions with a codeswitch between base and intensifier are only allowed if the intensifier is a lexical item rather than an affix. Furthermore, it is hypothesized that, if the Uniform Structure Principle is adhered to, synthetic intensifiers tend to occur if the ML is Spanish and are avoided if the ML is English. This would follow from the observation that the analytic strategy is the default within the English system, whereas both the synthetic and analytic strategies are salient within Spanish grammar.

Still, intralinguistic factors alone are expected not to suffice in explaining the attested patterns. In a study of intensifying strategies in youngsters’ Madrilenian Spanish, Roels (2024) finds all intralinguistic factors taken into consideration to be subordinate to the considered extralinguistic factors, which seem to be the best predictors of attested patterns. As such, it is investigated to what extent specific sociolinguistic variables6, which are shown to be of influence on bilinguals’ language use (Perani et al., 2003; Foote, 2010; Luo et al., 2010; Dewi & Setiadi, 2018; Kastenbaum et al., 2019), help to explain the intensifying patterns in the bilingual community under study. These include the language proficiency of the speakers in both English and Spanish, their age of acquisition of these languages, and their attitudes toward them. It is then expected that, if these factors are of influence, bilingual speakers increasingly opt for English or Spanish intensifying strategies if their self-reported proficiency, age of acquisition, and attitudes predominantly favor English or Spanish, respectively.

2.2.3. Hypotheses Regarding Convergence Effects on Intensifying Strategies

It is possible that the intensifying systems of English and Spanish in Miami show convergence effects, considering the effects of convergence on other linguistic phenomena in Miami (Lynch, 1999; P. M. Carter et al., 2020). Specifically, bilingual speakers may resort more frequently to intensifying strategies or even specific intensifiers shared between the two languages. In terms of strategies, it is expected that, if convergence is at play in Miami, analytic intensification is the preferred strategy in the observed bilingual speech as the strategy shared by both languages. Consequently, this means that Miami bilinguals, in case of convergence, are likely to use fewer synthetic intensifiers in their Spanish compared to monolingual speakers of related Spanish varieties.

However, if convergence effects apply, in their English, Miami bilinguals may actually use slightly more synthetic intensification than monolingual American English speakers. It is likely that this can, above all, be seen in a greater use of super- and the other intensifying prefixes in the English spoken by Miami bilinguals, as these specific items are shared between both languages. To test these convergence effects, the bilingual English and Spanish speech of Miami is compared to monolingual English and Spanish varieties.

3. Corpus and Methods

3.1. Data

In order to investigate the speech of Miami bilinguals, this study analyzes conversations from the Miami Corpus of BangorTalk (Deuchar et al., 2014)7. This corpus consists of 56 naturalistic spoken conversations between 83 Miami bilinguals, totaling approximately 35 hours of speech recorded in 2008 and 264,579 words, of which 166,665 (62.99%) are English, 90,844 (34.34%) are Spanish, and 7,070 (2.67%) cannot be classified as either English or Spanish.

For comparative purposes with respect to the research question regarding convergence, monolingual English and Spanish corpora are also required in order to compare the use of analytic and synthetic intensifiers in the speech variety of Miami to their use in monolingual English and Spanish varieties. Concretely, an analysis is run of data from the Santa Barbara Corpus (Du Bois et al., 2005)8, specifically a sample of 100,074 words consisting of all conversations with speakers identified as non-Hispanic from all southern U.S. states represented in the corpus, which are Florida, Alabama, Louisiana, Texas, New Mexico, and California. Furthermore, the Havana subcorpus of Ameresco is analyzed, which represents Cuban Spanish and consists of 102,890 words (Albelda Marco & Estellés Arguedas, 2016)9.

These corpora are chosen because they are closely comparable to the Miami Corpus in terms of type of speech, as all three contain naturalistic spontaneous conversations. Furthermore, the two monolingual corpora represent relevant varieties, American English and Cuban Spanish, respectively, to which to compare the variety of Miami, as these are likely closest to the English and Spanish spoken in Miami.

3.2. Parameters

To begin with, through a front-to-back reading of the transcripts from the three corpora, all constructions in the datasets containing an overt intensifier modifying a base of any word class are extracted for annotation. Additionally, constructions containing certain modifying adjectives are also included, namely those adjectives expressing a great size or a positive or negative evaluation (e.g., a big house, a good friend, a bad mood). These constructions, albeit arguably not intensifying constructions, are included because they form structural alternatives to the various intensifying affixes under study (e.g., a megahouse vs. a big house, a superfriend vs. a good friend, a moodzilla vs. a bad mood). Consequently, the proportion of synthetic forms uttered by speakers can only be properly investigated if these alternative formulations are included in the analysis. Therefore, in the analyses regarding the analytic–synthetic dichotomy where this proportion of synthetic forms is under consideration, these adjectival constructions are additionally included.

In a second phase, the extracted tokens are annotated10 for the intralinguistic and extralinguistic variables mentioned above, the latter of which are mostly derived from the sociolinguistic survey data included in the corpora. These variables, organized by the research questions that they are relevant to, are the following:

• Descriptive and identifying variables:

  ○

  Conversation and line in which construction is found;

  ○

  Code, gender, and age of speaker uttering construction;

  ○

  Code, gender, and age of hearer construction is uttered to.

• Variables relevant to the research question regarding influence of intralinguistic and extralinguistic factors on intensifying configuration:
  • Intralinguistic variables:

    ○

    Language of base (English vs. Spanish);

    ○

    Language of intensifier (English vs. Spanish);

    ○

    Semantic-pragmatic function of intensifier (quantitative vs. qualitative-evaluative);

    ○

    Strategy of intensifier (analytic vs. synthetic);

    ○

    Matrix language (English vs. Spanish).

  • Extralinguistic variables:

    ○

    Self-reported language proficiency11 (categorical, ranging from 0 ‘very low’ to 4 ‘very high’; of both English and Spanish, as well as the difference between both values);

    ○

    Age of acquisition (categorical, ranging from 1 ‘from birth’ to 5 ‘after adulthood’; of both English and Spanish, as well as the difference between both values);

    ○

    Language attitudes (continuous12, ranging from 1 ‘very negative’ to 5 ‘very positive’; of both English and Spanish, as well as the difference between both values).

• Variables relevant to the research question regarding convergence effects on intensifying strategies:

  ○

  Corpus in which construction is found (Miami, Santa Barbara, Havana);

  ○

  Strategy of intensifier (analytic vs. synthetic).

The attested intensifying constructions and their extralinguistic and intralinguistic properties are analyzed both qualitatively, in that the intensifying paradigm of the community is described by discussing the various attested intensifying patterns, and quantitatively, in that the frequencies of certain patterns over others and the relationship of these frequencies to relevant variables are tested statistically. A relevant test, namely either a chi-square or Fisher’s exact test, is performed for every variable separately. The results of the analyses are found in the following section.

4. Results

In the following section, the results are described and analyzed statistically. First, a descriptive overview of the attested constructions is given, followed by more detailed results, as well as appropriate statistical analyses, in order to answer the respective research questions.

4.1. Formal Analysis of Intensification Strategies

First of all, a general description is provided of the grammar of intensification as found in Miami. From the entire Miami Corpus, a total of 1,737 constructions are extracted. Of these, 1,282 are true intensifying constructions, while the rest constitute the aforementioned adjectival constructions (e.g., a big house, a good friend, a bad mood).

Table 3. Frequencies of true intensifying and adjectival constructions in both languages.

Language	True Intensifying		Adjectival		Total
	freqabs	%	freqabs	%	freqabs	%
unilingual English	902	70.36%	324	71.21%	1,226	70.58%
unilingual Spanish	369	28.78%	126	27.69%	495	28.50%
bilingual	11	0.86%	5	1.10%	16	0.92%
total	1,282		455		1,737

shows the division of the found constructions according to these categories and the language in which they appear.

Immediately striking is the very low frequency of bilingual intensifying constructions (freq_abs = 11, 0.86%), where an English base is modified by a Spanish intensifier or vice versa. This is apparently a dispreferred pattern in the English–Spanish speech of Miami. The 11 constructions are discussed in more detail in Section 4.2.1. The data further show that there are many more English than Spanish intensifying constructions. At first sight, this can be explained by the fact that English is much more dominant in the corpus than Spanish.

Table 4. Frequencies of unilingual intensifying constructions by word class of base.

Word Class	Unilingual English		Unilingual Spanish
	freqabs	freqnorm	freqabs	freqnorm
adjective	624	37.44	226	24.88
verb	97	5.82	31	3.41
adverb	69	4.14	57	6.27
noun	45	2.70	47	5.17
pronoun	32	1.92	5	0.55
determiner	22	1.32	0	0.00
prepositional phrase	10	0.60	3	0.33
subordinate clause	3	0.18	0	0.00
total	902	54.12	369	40.62

, however, gives more insight into this. Here, the division is shown of all unilingual true intensifying constructions by the word class of the base, as defined in Section 2.1.1, with normalized frequencies per 10,000 English or Spanish words between brackets. The total normalized frequencies show that, even when controlling for the overall presence of English and Spanish in the corpus, there are relatively more English (freq_norm = 54.12) than Spanish (freq_norm = 40.62) intensifiers in the corpus.

The overall greater productivity of English intensifiers extends across all possible types of bases, with the exception of adverbial and nominal bases, which are intensified relatively more often in Spanish (6.27 and 5.17 per 10,000 words, respectively) than in English (4.14 and 2.70 per 10,000 words, respectively). For the nominal bases, this can be explained by the fact that the Spanish number also includes instances of the augmentative suffixes, which English generally lacks; indeed, 37 of the intensified Spanish nouns (78.72%) constitute instances of one of these suffixes (e.g., un molote ‘a big riot’).

Below, some examples of unilingual intensifying constructions found in the corpus are shown. First, examples are given of English intensifying constructions involving an English adjective (12a), an adverb (12b), a noun (12c), and a verb (12d), followed by examples of Spanish analytic intensifiers used on a Spanish adjective (13a), an adverb (13b), a noun (13c), and a verb (13d) and, finally, examples of Spanish synthetic intensifiers used on and adjective (14a), an adverb (14b), and a noun (14c).

(12)	a.	Yeah no but you know it’s funny, but Hawaii, Hawaii is beautiful, very beautiful. (Miami, zeledon2)
	b.	This is the one I got in quick as fuck, but at the end I didn’t care that much. (Miami, sastre2)
	c.	The invitations so far as I have imagined them are in English, but then I was thinking everyone in the fucking family speaks Spanish. (Miami, herring9)
	d.	She hasn’t given me back my grade on the presentation and it really bothers me that Peter got a, the same thing that I got. (Miami, herring6)

(13)	a.	Están	museo	de	las	ciencia-s,	museo	de	la	luz
		be.3pl	museum	of	the	science-pl	museum	of	the	light
		y	otros	dos,	debe	ser	un	complejo	muy	grande.
		and	other	two	must.3sg	be	a	complex	very	big
		‘There’s the science museum, the museum of light and two more, it must be a very big complex.’ (Miami, herring5)

	b.	Sale	desde	bien	temprano	y	llega	bien	tarde.
		leave.3sg	from	very	early	and	arrive.3sg	very	late
		‘He goes very early and returns very late.’ (Miami, herring11)

	c.	Tengo	una	hambre	de	mil	diablo-s,	te	cuento.
		have.1sg	a	hunger	of	thousand	devil-pl	you	tell.1sg
		‘I tell you, I’m hungry like a thousand devils.’ (Miami, herring16)

	d.	Siempre	me	gustaba	mucho	la	odontología,	siempre
		always	me	please.3sg.impf	much	the	dentistry	always
		estaba	muy	involucrada	en	la	odontología.
		be.1sg.impf	very	involved	in	the	dentistry
		‘I always really liked the dentist’s, I was always very involved in dentistry.’ (Miami, zeledon1)

(14)	a.	Te	cagás	de	la	risa,	divertid-ísimo	buen-ísimo	es	él.
		you	shit.2sg	of	the	laugh	funny-sup	good-sup	be.3sg	he
		‘You’ll laugh your ass off, he’s really funny.’ (Miami, herring2)

b.

Qué

cagada,

creo

que

queda

lej-ísimos

esa

verga,

what

crap

believe.1sg

that

stay.3sg

far-sup

that

dick

me

queda

super-lejos.

me

stay.3sg

ints-far

‘What crap, I think that shit is too far, it’s too far for me.’ (Miami, herring12)

	c.	¿También	no	hubo	una,	una,	un	mol-ote	por	el	precio
		also	not	be.3sg.pst	a	a	a	mass-aug	for	the	price
		de	las	tortilla-s?
		of	the	tortilla-pl
		‘Also wasn’t there a big riot about the price of tortillas?’ (Miami, herring5)

Next,

Table 5. Frequencies of unilingual intensifying constructions by strategy of choice.

Strategy	Unilingual English		Unilingual Spanish		Total
	freqabs	%	freqabs	%	freqabs	%
analytic	1,016	95.67%	359	78.73%	1,375	90.58%
synthetic	46	4.33%	97	21.27%	143	9.42%
total	1,062		456		1,518

takes a broader perspective and shows the division of unilingual constructions in both languages across the two strategies, analytic and synthetic. Note that the data shown here include only those constructions with adjectives, adverbs, and nouns as bases, because these three word classes most clearly show the alternation between analytic and synthetic forms in the intensifiers they take (e.g., very cheap vs. cheapass, very quickly vs. superquickly, a big house vs. a megahouse). It can be seen that synthetic forms make up a much larger proportion of the Spanish intensifiers (21.27%) compared to the English ones (4.33%).

These are the results that are to be compared to those of the monolingual communities in Section 4.3. First, however, the next subsection discusses in some more detail the codeswitched intensifying constructions observed in the corpus.

4.2. Quantitative Analysis of Intensification Strategies in Bilingual Contexts

4.2.1. Analysis of Impact of Intralinguistic Parameters on Intensification Strategies

This subsection investigates the effects that various intralinguistic and extralinguistic factors have on speakers’ intensifying strategies. First, the attention is turned to intensifying constructions in bilingual contexts, so any intensifying construction situated in a clause in which codeswitching occurs. One should note that, in the Miami Corpus, very few intensifying constructions are attested in bilingual contexts, namely 109 or 8.50% of the overall tokens of intensification. In these bilingual contexts, the question arises whether, when both Spanish and English are activated, bilinguals tend to relate the use of intensifiers in one or the other language to semantically specialized meanings (Blom & Gumperz, 1972; Aaron, 2004). Specifically, previous research suggests that English–Spanish bilinguals may prefer English to discuss objective notions and Spanish to express more affective statements (Vinagre Laranjeira, 2005; Betti & Enghels, 2018). If this observation also holds true for intensification, Miami bilinguals should favor English for quantitative and Spanish for qualitative-evaluative intensification.

Table 6. Frequencies of intensifiers in bilingual contexts by semantic-pragmatic function.

Function	English Intensifiers	Spanish Intensifiers
quantitative	62	40
qualitative-evaluative	5	2
total	67	42

shows the frequencies of English and Spanish intensifiers in bilingual contexts expressing quantitative and qualitative-evaluative values. One can see here that the quantitative function is overall much more frequent than the qualitative-evaluative function. When testing the relationship between the function and language of the intensifier, it is found that there is no statistically significant relationship between these two variables (p = .705)13.

These findings imply that the semantic-pragmatic function of an intensifier does not have an effect on speakers’ choices to express this intensifier in either English or Spanish. Indeed, the data show many cases of quantitative and qualitative-evaluative intensifiers in both English and Spanish. Below, (15a) shows an example of an English construction with a quantitative function, while (15b) shows a Spanish one. Next, (16a) shows an English construction and (16b) a Spanish construction, both with a qualitative-evaluative function. Note that all examples below constitute instances of adjectival constructions rather than true intensifying constructions, as true intensifiers, such as the intensifying affixes, are somewhat rare in bilingual contexts.

(15)	a.	That,	that real,	real pig,	los,	los,	los	with a big	belly?
		that	that real	real pig	the	the	the	with a big	belly
		‘A real pig, with a big belly?’ (Miami, sastre1)

b.

You

know el,

el

bad guy,

the bad guy in part two,

el

que

you

know the

the

bad guy

the bad guy in part two

the

who

tiene,

el,

el

cubano

que

tiene

pelo

largo?

have.3sg

the

the

Cuban

who

have.3sg

hair

long

‘You know the bad guy in part two, the one who has, the Cuban who has long hair?’ (Miami, herring7)

(16)	a.	Así	que	si	escuchas	de	algún	good	deal,	because	en
		like.this	that	if	listen.2sg	of	any	good	deal	because	in
		el	dealer donde	él	está	hay	carro-s.
		the	dealer where	he	be.3sg	be.3sg	car-pl
		‘So if you hear of any good deal, because at the dealer where he is there are cars.’ (Miami, zeledon5)

	b.	Él	puede	trabajar	en	la	biblioteca	también,	sería
		he	can.3sg	work	in	the	library	also	be.3sg.cond
		un	buen	trabajo	then.
		a	good	work	then
		‘He can work at the library too, it could be a good work then.’ (Miami, sastre9)

Furthermore, the influence of certain structural constraints is to be considered. The first of these to consider is the Free Morpheme Constraint, which postulates that a codeswitch should not occur between a stem and a morpheme bound to this stem. In order to investigate this, the focus is narrowed down to only the bilingual constructions. As explained above, only 11 bilingual intensifying constructions, with either an English base intensified by a Spanish intensifier or vice versa, are found in the data. Below,

Table 7. Frequencies of bilingual intensifying constructions by word class of base.

Word Class	English base + Spanish ints	Spanish base + English ints
adjective	7	2
noun	0	1
determiner	0	1
total	7	4

shows the attested patterns observed for these constructions in terms of the word class of the base and the language in which base and intensifier occur. Below that, the eleven constructions in question are shown, first the seven English bases with Spanish intensifiers (17a–f), followed by the four Spanish bases with English intensifiers (18a–d). The data show that, despite the low frequency of bilingual constructions, the attested patterns are moderately diverse, although there are more with English than Spanish bases and most contain an adjective as base. The majority of these adjectives are intensified by adverbs, all of which are among the most prototypical intensifying adverbs, while two are instead intensified by the prepositional phrase de lo más ‘very’.

(17)	a.	Yeah	está	muy	modern	y	nuevo.
		yeah	be.3sg	very	modern	and	new
		‘Yeah it’s very modern and new.’ (Miami, sastre8)

	b.	Él	estaba	muy	busy.
		he	be.3sg.impf	very	busy
		‘He was very busy.’ (Miami, sastre11)

	c.	¿Paul	es	muy,	muy	hotheaded?
		Paul	be.3sg	very	very	hotheaded
		‘Is Paul very hotheaded?’ (Miami, herring6)

	d.	Pero	después	cuando	llueve	se	pone	muy	slippery.
		but	after	when	rain.3sg	itself	put.3sg	very	slippery
		‘But after that it gets very slippery.’ (Miami, herring14)

	e.	Es	así	tan	rude	con	mis	amigo-s,	dios	mío.
		be.3sg	like.this	so	rude	with	my	friend-pl	god	my
		‘She is so rude with my friends anyway, good god.’ (Miami, herring16)

	f.	¿Tú	viste?	Está	de	lo	más	clean,	está	de
		you	see.2sg.pst	be.3sg	of	the	most	clean	be.3sg	of
		lo	más	clean.
		the	most	clean
		‘Did you see it? It’s very clean, it’s very clean.’ (Miami, maria16)

(18)	a.	We	were together for four years and he was very	machista.
		we	were together for four years and he was very	male.chauvinistic
		‘We were together for four years and he was very male chauvinistic.’ (Miami, zeledon2)

	b.	You	know the president of the club,	no		tú	sabes,		he’s real
		you	know the president of the club	no		you	know.2sg		he’s real
		político.
		political
		‘You know the president of the club, no you know, he’s really political.’ (Miami, sastre6)

	c.	So	I throw out while we go in and it’s a freaking	botánica.
		so	I throw out while we go in and it’s a freaking	garden
		‘So I throw out while we go in and it’s a freaking garden.’ (Miami, sastre7)

	d.	Ah	that whole area you would not catch me dead,						it could be fucking
		ah	that whole area you would not catch me dead						it could be fucking
		ahí		todo	el	mundo.
		there		all	the	world
		‘Ah that whole area you would not catch me dead, it could be fucking everyone there.’ (Miami, sastre4)

It should be noted that all of these constructions are instances of insertion (Muysken, 1997), where a base in a particular language is inserted into a context otherwise predominantly featuring the other language. In all cases, the language of the intensifier additionally aligns with the ML. One can also observe that all of these constructions feature analytic intensifiers. Based on this, the Free Morpheme Constraint appears to be adhered to in the data, as no synthetic intensifying constructions are attested in which a codeswitch occurs.

At this point in the analysis, it is of interest to investigate the influence of the Uniform Structure Principle, which predicts that particular grammatical features in a bilingual context align with the grammar of the ML. If this applies to the data discussed here, an English ML is expected to show a dispreference for synthetic intensification.

Table 8. Frequencies of intensifiers in English-ML and Spanish-ML bilingual contexts by strategy of choice.

Strategy	English ML	Spanish ML
analytic	74	52
synthetic	1	11
total	75	63

shows the division of intensifying strategies in bilingual contexts. As can be seen here, only one synthetic intensifying constructions occurs in an English-ML context. Conversely, there are some more synthetic constructions to be found in a Spanish-ML context, which is expected to be more lenient in terms of possible intensifying strategies, due to both strategies being more productive within the Spanish system. When statistically testing this relationship, it is found that the ML and strategy of choice are significantly related to each other, additionally displaying a moderately strong effect size (χ² = 11.22, p < .001, V = .29).

From the results presented in this subsection, it can be seen that bilinguals’ intensifying strategies are not significantly related to the semantic-pragmatic function of the intensifiers, but that bilinguals’ choices in this regard appear to be constrained by the Free Morpheme Constraint and the Uniform Structure Principle, as the data show a dispreference toward synthetic intensifying constructions with a switch between the intensifier and either the base or the ML.

4.2.2. Analysis of Impact of Extralinguistic Parameters on Intensification Strategies

The attention is next shifted to extralinguistic variables in explaining bilingual speakers’ intensifying strategies. Some factors identified in previous research that are likely influential in this respect are various of the sociolinguistic characteristics of the speakers, most importantly their language proficiency, age of acquisition, and language attitudes (Foote, 2010; Parafita Couto et al., 2014; Dewi & Setiadi, 2018). More precisely, if an individual’s proficiency, language acquisition, or attitudes of English and Spanish differ significantly, this variation is expected to influence their use of intensifiers in bilingual contexts. Specifically, speakers are likely to employ more English or Spanish intensifiers depending which language they demonstrate higher proficiency in, earlier acquisition of, or more positive attitudes toward. Consequently, the study examines to which extent speakers use more English intensifiers when they report to have greater proficiency in English than in Spanish, acquire English earlier than Spanish or hold a more positive overall attitude toward English. Conversely, it analyzes to what degree speakers employ more Spanish intensifiers when they report higher proficiency in Spanish, acquire it earlier than English or have a more favorable attitude toward Spanish.

First of all, the influence of the difference between speakers’ self-reported English and Spanish language proficiency is tested.

Table 9. Number of participants and intensifiers per difference score between self-reported language proficiency values.

Difference in Proficiency	Participants	English Intensifiers	Spanish Intensifiers
English 2 > Spanish	7	8	8
English 1 > Spanish	7	0	0
English = Spanish	61	56	33
English < 1 Spanish	7	3	1
English < 2 Spanish	1	0	0
total	83	67	42

shows the division of participants from the Miami Corpus by the difference between their proficiency scores in English and Spanish and the number of uttered English and Spanish intensifiers in bilingual contexts divided across these groups14. As can be seen here, most participants, namely 61 out of 83, report equal proficiency in both languages. When testing the influence of difference in proficiency on the language of choice of the intensifying construction using a Fisher’s exact test, no statistically significant relationship is found (p = .581), which means that bilinguals appear not to use more intensifiers in the language that they have a higher proficiency in. Note that one possible explanation of these results concerns the fact that self-reported proficiency is used in the analysis instead of a more objective measure of proficiency (see Note 11).

Second, speakers’ age of acquisition is considered as a factor.

Table 10. Number of participants and intensifiers per difference score between age-of-acquisition values.

Difference in AoA	Participants	English Intensifiers	Spanish Intensifiers
English 4 > Spanish	11	5	3
English 3 > Spanish	8	4	1
English 2 > Spanish	26	31	19
English 1 > Spanish	12	3	2
English = Spanish	22	21	17
English < 1 Spanish	2	1	0
English < 2 Spanish	2	0	0
total	83	67	42

shows the number of speakers from the corpus for each difference value between individual speakers’ age-of-acquisition scores and the number of intensifiers uttered in bilingual contexts divided by the speakers’ age-of-acquisition differences (see Note 14). Note that the age-of-acquisition scores are measured on a five-point scale, with each of these five categories representing a different age range, ranging from acquisition from birth to acquisition into adulthood. It can be seen that the majority of speakers have acquired Spanish earlier than English, with only a few acquiring English before Spanish. When testing whether a difference in age of acquisition stands in relation to a different usage of English and Spanish intensifiers with a Fisher’s exact test, it is found that there is no statistically significant relationship between the two (p = .937). This suggests that speakers do not prefer to intensify in the language that they acquire first.

Finally, speakers’ language attitudes are considered.

Table 11. Number of participants and intensifiers per difference score between language attitude values.

Difference in Attitude	Participants	English Intensifiers	Spanish Intensifiers
English 2+ > Spanish	1	1	0
English 1–2 > Spanish	5	1	3
English 0–1 > Spanish	18	20	7
English = Spanish	19	23	10
English < 0–1 Spanish	34	18	22
English < 1–2 Spanish	6	4	0
total	83	67	42

shows the number of participants from the corpus sorted by the difference between their attitudes toward English and Spanish and the division of intensifiers uttered in bilingual contexts across these different groups (see Note 14). Note that the difference values are a little different than for proficiency and age of acquisition, as the language attitude values behave more like continuous variables. When using a Fisher’s exact test to test the relationship between speakers’ differences in attitudes and the language in which they express their intensifiers, a statistically significant effect is found (p = .017), with speakers using more English intensifiers if they evaluate English more positively than Spanish and vice versa. Additional testing reveals a moderately strong effect size (V = .34).

As can be seen from the analyses above, two of the three variables under consideration do not display a statistically significant relationship with language of choice. Of the three, only language attitudes appear to be an indicator of the language in which bilinguals prefer to express their intensifiers in a bilingual context.

4.3. Quantitative Analyses of Convergence Effects on Intensification Strategies

Building on the previous observations, this subsection examines the hypotheses related to convergence and, more concretely, the proportion of synthetic or analytic intensifiers in Miami compared to monolingual communities. First up, Miami is compared to data from the Santa Barbara Corpus so as to make a comparison between Miami English and American English in general. It is expected that, in case of convergence with the Spanish intensifying paradigm, more synthetic intensifiers are used in the Miami Corpus than in the Santa Barbara Corpus. In

Table 12. Frequencies of English intensifying constructions in Miami and Santa Barbara Corpus by strategy of choice.

Strategy	Miami		Santa Barbara
	freqabs	%	freqabs	%
analytic	1,016	95.67%	791	99.75%
synthetic	46	4.33%	2	0.25%
total	1,062		793

, a comparison is shown between the proportion of synthetic intensifiers in the Miami Corpus and the Santa Barbara Corpus. It can be seen here that, indeed, more synthetic intensifiers are used in Miami English compared to American English in general. This difference is statistically significant, but has a rather weak effect size (χ² = 29.97, p < .001, V = .13). As such, even though there is no strong relationship to be found, the attestation of a significant effect may hint at some slight convergence effects.

Secondly,

Table 13. Frequencies of Spanish intensifying constructions in Miami and Havana by strategy of choice.

Strategy	Miami		Havana
	freqabs	%	freqabs	%
analytic	359	78.73%	316	66.67%
synthetic	97	21.27%	158	33.33%
total	456		474

compares the proportions of synthetic intensifiers as found in Miami and Havana. One can clearly see here that far fewer synthetic forms are used in Miami than in Havana. This difference is statistically significant, although it has a weak effect size (χ² = 16.99, p < .001, V = .14). These findings may thus hint at some beginning, albeit slight, convergence effects in Miami Spanish, with speakers using fewer synthetic intensifiers in favor of analytic ones, which is more in accordance with English.

5. Discussion and Conclusions

This paper examined the phenomenon of intensification in the speech of the bilingual community of Miami, a city characterized by widespread English–Spanish bilingualism. In this context, contact effects such as codeswitching and convergence can be observed. A central question addressed in this study is how intensification manifests in this bilingual environment. Notably, intensification forms a conflict site between the grammars of English and Spanish, with Spanish having a much richer paradigm of intensifying affixes, especially because of its suffixes, while English relies for a larger part on lexical items to intensify. Additionally, intensification is a phenomenon that can convey a variety of both objective and affective notions. It is of interest to explore how bilingual speakers express these different notions when they have such a rich variety of options at their disposal to intensify, being able to derive from the intensifying systems of two different languages.

This study examined the patterns underlying speakers’ choices among the various available intensifying strategies, guided by two research questions. First, it investigated to what extent the configuration of Miami bilinguals’ intensifying constructions is influenced by the semantic-pragmatic function of the intensifier and relevant structural constraints and theories, as well as speakers’ self-reported proficiency, age of acquisition, and attitudes of English and Spanish. Second, it examined to what extent Miami bilinguals’ intensifying paradigm displays convergence effects as to the employed strategies and their frequencies when compared to monolingual English and Spanish varieties. To address these questions, corpus data from three sources were analyzed, specifically the bilingual Miami Corpus and two monolingual corpora, namely the Santa Barbara Corpus for English and the Havana subcorpus of Ameresco for Spanish.

Regarding the results, the data revealed a wide range of intensifying constructions in both English and Spanish, with a notable preference for English intensifiers over Spanish ones. However, the majority of these constructions are unilingual, while bilingual constructions, where a codeswitch occurs between the base and the intensifier, are exceedingly rare. This suggests that such bilingual patterns are dispreferred in the English–Spanish speech of Miami.

Among the few bilingual constructions observed, no synthetic intensifiers were found, aligning with the predictions of the Free Morpheme Constraint. Expanding the analysis to include both unilingual and bilingual intensifying constructions in bilingual contexts, the findings indicate an overall scarcity of synthetic intensifiers in these contexts, particularly when English serves as the matrix language. This pattern is consistent with the Uniform Structure Principle. Also in these bilingual contexts, no statistically significant relationship was found between the semantic-pragmatic function of an intensifier and the language in which it is expressed. Consequently, the analysis shifted from intralinguistic to extralinguistic factors to account for the observed patterns. Among the extralinguistic variables examined (self-reported language proficiency, age of acquisition, and language attitudes toward English and Spanish) only language attitudes were found to significantly influence the language choice for intensifiers in bilingual contexts.

The absence of a relationship between semantic-pragmatic function and language choice is not necessarily unexpected. It is possible that intensification, as a linguistic phenomenon, does not conform to the dichotomy often observed in bilingual speech, where English is associated with objective expressions and Spanish with subjective ones (Vinagre Laranjeira, 2005). More broadly, previous research already demonstrates that extralinguistic factors tend to be stronger predictors of speakers’ intensifying strategies than intralinguistic factors (Roels, 2024). When considering these extralinguistic factors, it is interesting to observe that two of the three included variables show no significant relationship with language of choice, while these variables do show significant effects on bilingual speech in previous research. The apparent greater importance of language attitudes over proficiency and age of acquisition, as observed in this study, warrants further exploration. This finding suggests that bilingual speakers’ language attitudes may exert a more substantial influence than previously assumed, at least within specific contexts such as the one examined here.

As for the second research question, the findings suggest that Miami bilinguals use significantly more synthetic intensifiers in their English compared to monolingual English speakers, while employing fewer synthetic intensifiers in their Spanish than monolingual Spanish speakers. These patterns may indicate the presence of convergence effects in Miami, although these effects appear to be minimal or in an early stage of development.

By investigating intensification within the bilingual community of Miami, the current study offers additional insights to existing research on bilingualism and language contact. However, some of the findings are based on a limited number of observations, meaning that not all conclusions can be drawn with equal certainty. Future research could reinforce these findings by analyzing a larger set of bilingual data. Additionally, further studies could explore the same phenomenon in other English–Spanish bilingual communities or in bilingual communities with different language pairings. In conclusion, this study offers a valuable and innovative contribution to research on expressive linguistic phenomena, such as intensification, within bilingual contexts like the one of Miami.