Multilingualism, Translanguaging and Baloney


In my post on Multilingualism and Translanguaging I made an effort to put a positive spin on work being done in the areas of multilingualism, translanguaging, raciolinguistics and Disability Critical Race Theory. I emphasised that this work is unified in its radical objective of challenging current language learning and language teaching practices and replacing them with “new configurations, which allow for the emergence of discourses and voices that are currently deliberately repressed by the status quo”. My argument was that while there are good reasons to support the general thrust of arguments in favor of multilingualism, there is no need for enthusiastic followers of this work to accuse those pursuing scientific research in psycholinguistics of imposing a “positivist paradigm” on applied linguistics. I suggested that the two domains of sociolinguistics and psycholinguistics need not lock horns in a culture war, just so long as those in the sociolinguistic camp don’t disappear down a relativist, socio-cultural hole where appeals to logic and empirical evidence are entirely abandoned.

Here, I take a closer, more critical look at the work of García, Flores and Rosa. My criticism is motivated by an attempt to persuade my chum Kevin Gregg that I haven’t completely lost the plot, and to persuade other readers that García, Flores and Rosa indulge in theoretical speculations, inspired by an extreme relativist epistemology, which result in little more than obscurantist baloney that gets nobody anywhere.

My Position

Let me begin by summarizing what I, from my critical rationalist perspective, acknowledge:

  • English acts as a lingua franca and as a powerful tool to protect and promote the interests of a capitalist class.
  • In the global ELT industry, teaching is informed by the monolingual fallacy, the native speaker fallacy and the subtractive fallacy (Phillipson, 2018).   
  • The ways in which English is privileged in education systems needs critical scrutiny, and policies that strengthen linguistic diversity are needed to counteract linguistic imperialism.
  • Translanguaging affirms bilinguals’ fluent languaging practices and aims to legitimise hybrid language uses.
  • ELT must generate translanguaging spaces where practices which explore the full range of users’ repertoires in creative and transformative ways are encouraged.
  • Translanguaging classroom practices can disrupt subtractive approaches to language education and deficit language policies.
  • Racism permeates ELT. It results in expecting language-minoritized students to model their linguistic practices on inapproriate white speaker norms.

Additive Bilingualism

To expose the baloney which I think characterises the work of García, Flores and Rosa, I use Cunmmins’ (2017) article, where he defends the construct of additive bilingualism and additive approaches to language education against criticisms made by them. I make comments that Cummins will certainly disagree with.

Cummins starts with Baker and Prys Jones’ (1998) definitions:

Additive Bilingualism: A situation where a second language is learnt by an individual or group without detracting from the maintenance and development of the first language. A situation where a second language adds to, rather than replaces the first language. (p. 698)

Subtractive Bilingualism: A situation in which a second language is learnt at the expense of the first language, and gradually replaces the first language (e.g. inmigrants (sic) to a country or minority language pupils in submersion education) (p. 706).”

 Cummins points out that although the additive/subtractive distinction was originally formulated as a psycholinguistic construct, it “rapidly evolved to intersect with issues of education equity and societal power relations”. His later definition is that additive bilingualism seeks to “help students add a second language (L2) while continuing to develop academic skills in their home language (L1)”.

Attacks on Cummins’ view begin with García’s (2009) claim that additive bilingualism positions bilingualism as the sum of two monolingualisms. She advocates replacing this mistaken view with the construct of translanguaging, which can act as both a description of the dynamic integrated linguistic practices of bilingual and multilingual students and as a pedagogical approach. Flores and Rosa (2015) build on García’s criticism by arguing that additive approaches interpret the linguistic practices of bilinguals through a monolingual framework where “discourses of linguistic appropriateness fueled by raciolinguistic ideologies” remain unchallenged. As a result, an additative approach “marginalizes” the “fluid linguistic practices” of bilingual students, Standard English learners, and heritage language learners (the three “communities” that Flores and Rosa (2015) deal with).

Cummins replies that García’s and Flores & Rosa’s conceptualizations of additive bilingualism “load the construct with extraneous conceptual baggage that is not intrinsic to its basic meaning”. Additive bilingualism, he argues, does not imply that bilingualism is the sum of two monolingualisms, and additive approaches to language education are as radical – and more coherent – than those of García and Flores & Rosa.

… the construct of additive bilingualism has evolved from its psycholinguistic roots to reference a set of education practices and initiatives that challenge the operation of coercive power structures. These power structures have historically excluded minoritized students’ L1 from schooling with the goal of replacing it with the L2. Extensive research carried out within the context of the additive bilingualism construct has demonstrated that minoritized students’ L1 can be promoted through bilingual education programs at no cost to students’ academic development in the L2″ (Cummins, 2017, p. 408).

So what is the “extraneous conceptual baggage” that has been loaded onto Cummins’ construct of additive bilingualism? It is, precisely,

the assumption that the construct of additive bilingualism necessarily entails distinct language systems rather than functioning as an integrated system” (Cummins, 2017, p. 411).

To be clear, the extraneous conceptual baggage resides in García’s rejection of the assumption that the categories “first language and second language” are meaningful. García insists that first language and second language are not actually distinct language systems, thus denying that Mandarin Chinese, Hindi, Spanish and English, for example, are distinct language systems. Given this preposterous claim, it should not come as a surprise that García and Wei (2014) take the extra step into Humpy Dumpty land by claiming that the construct of language/languages itself is illegitimate. According to them, the translanguaging construct demands that bilingual students’ language practices must not be separated into home language and school language. It follows that the concept of transfer must be shed in favor of “a conceptualization of integration of language practices in the person of the learner” (García & Wei, 2014, p. 80).

So …..

One consequence of the assertion that languages don’t exist is that it is meaningless to talk about transfer from one language to another. It also means, as Cummins points out, that a child who says “My home language is English, but my school language is French” is making an ideologically-charged false utterance and that any source which refers to and provides information about the 7,106 languages and dialects that humanity has generated is similarly epistemologically wrong-headed. Cummins comments:

“The claim that languages exist as social constructions but have no legitimacy “in reality” raises the issue of what “reality” and “social construction” are. Also unclear is the meaning of the claim that languages don’t exist as linguistic entities but do exist in the social world” (Cummins, 2017, p. 414).

Those of us who, unlike Cummins, question the merits of adopting an exclusively socio-cultural view of language and language learning will go further than that. The claim that “languages exist as social constructions” itself needs clarifying. What kind of claim is it? Does it base itself on the work of Vygotsky, Lantolf, or what? Unless it’s properly explained, it’s so much blather. But still, we can agree with Cummins that the claim that languages themselves have no legitimacy “in reality” raises the issue of what the hell they’re talking about. What “reality” are they referring to? Surely, according to any decent relativist epistemology, no stable “reality” exists. And what does seeing language as a “social construction” mean in the context of a “framework” where languages don’t exist?

Rather than dismissing such crap as obvious nonsense, Cummins bends over backwards to reconcile himself with his critics. He himself adopts a socio-cultural approach where a relativist epistemology is often in evidence, but where appeals to empirical evidence are nevertheless made. His article ends with an attempt to reconcile differences with García, Flores and Rosa by proposing a synthesis of perspectives that replaces the term “additive bilingualism” with “active bilingualism”. Good luck with that, Jim. It seems obvious to me that this overture will be roundly rejected. García, Flores and Rosa are committed to an agenda that is characterised by its aggressive rejection of any compromise. A relativist epistemology is coupled with hopelessly undefined constructs and a disdain for empirical evidence that makes it impervious to criticism and encourages the worst kind of cult following.

There’s more

In a reply to Cummins (2017) Flores says that Cummins and he are using different theories of language.

“For Cummins, language is a set of disembodied features that exist as separate from the people who use them and can be objectively documented by researchers. I reject this premise and instead believe that language is embodied in ways such that the social status of the speaker can impact how their language practices are taken up by the listener, which could include researchers”.

Cummins’s view of language is not fairly described by Flores, who himself nowhere gives his own view. To say that “language is embodied in ways such that bla bla bla” is not to explain any theory of language. Flores talks a lot about “academic language” but refuses to say what his theory of language is.

In a recent tweet, Flores says:

The tendency is to start conversations about language in the classroom around how to differentiate for ELs. This is the wrong starting point. You can’t differentiate effectively if you don’t have a understanding of what language is and how it is embedded in relations of power.

I replied

Please can you tell me where I can find your answer to the question “What is language?”     

And answer came there none. Does Flores agree with García’s nonsensical statement that languages exist as social constructions but have no legitimacy “in reality”?

Other tweets from Flores illustrate his stance. For example:

If we are going to psychologize racial oppression it seems to me that the white kid who believes the white doll is more beautiful than the Black doll has more psychological damage than the Black kid who believes the same thing. Their beliefs will also do more damage in the world.

For the record, I don’t think psychologizing racial oppression is generally a good idea. I’m just saying it we are going to do it at least do it accurately. (emojoi of man shrugging).      

I leave those versed in critical discourse analysis to interpret these remarks.

There remains the question of how the works of García, Flores & Rosa affect teaching practice. What do they offer in terms of how to improve language teaching? Cummins (2017) argues that Flores and Rosa do not address the instructional implications of their theoretical claims about raciolinguistic ideologies.

The relevance of this construct for teaching minoritized students will emerge in much more powerful ways when the authors address the apparent contradictions I identify and specify alternative instructional approaches to expand students’ academic registers in place of the approaches advocated by Delpit (1988, 2006) and Olsen (2010), which they reject. Additionally, it would be helpful to address the apparent inconsistency between the endorsement of additive approaches in the work of Bartlett and García (2011) and the rejection of additive approaches in the Flores and Rosa analysis (Cummins, 2017, p. 420).

Amen to that. Cummins’ article has the title “Teaching Minoritized Students: Are Additive Approaches Legitimate?”. His prime concern is to improve the teaching of minoritised students and he’s worked hard to provide answers to that question. I see very little of practical use in the assorted works of García, Flores and Rosa. Pues, I think it’s all mucho noise and pocas nueces.


Bartlett, L., & García, O. (2011). Additive schooling in subtractive times: Bilingual education and Dominican immigrant youth in the Heights. Nashville, TN: Vanderbilt University Press.

Cummins, J. (2017). Teaching Minoritized Students: Are Additive Approaches Legitimate? Harvard Educational Review, 87, 3, 404-425.

Flores, N., & Rosa, J. (2015). Undoing Appropriateness: Raciolinguistic Ideologies and Language Diversity in Education. Harvard Educational Review, 85, 2, 149–171.

García, O. (2009). Bilingual education in the 21st century: A global perspective. Malden, MA: Wiley-Blackwell.

Garcia, O. & Wei, L. (2014) Translanguaging: Language, Bilingualism, and Education.  NY: Palgrave MacMillan.

Migliarini, V., & Stinson, C. (2021). Disability Critical Race Theory Solidarity Approach to Transform Pedagogy and Classroom Culture in TESOL. TESOL Journal, 55, 3, 708-718.

Phillipson, R. (2018) Linguistic Imperialism. Downloaded 28 Oct. 2021 from

Rosa, J., & Flores, N. (2017). Unsettling race and language: Toward a raciolinguistic perspective. Language in Society, 46, 5, 621-647.

13 thoughts on “Multilingualism, Translanguaging and Baloney

  1. Hi Geoff
    One quick point re ” any source which refers to and provides information about the 7,106 languages and dialects that humanity has generated is similarly epistemologically wrong-headed.”

    the Ethnologue database assumes languages are discrete and countable so you can see the circularity here and this is highlighted futher by the fact that named languages (i.e. what lay people call a language) number in the tens of thousands.

  2. Good heavens, Geoff, I never thought you were losing the plot; I just was expressing my astonishment that you have the energy to plow through this horseshit and actually deal with it. I’ll just make a couple of points here;
    1) Have Garcia and co-conspirators ever–EVER!–proposed a single empirical finding that their perspective/position/argument/bollocks has led to? (You actually don’t have to answer that, since we know the answer.)
    Actually, that’s a question not a point, but still. And of course it’s not a real question.
    2) What gets me even more than the intellectual wankery of these people is the pretension of left-wing virtue. As if by maundering on about embodiment and minority languages is somehow doing something–anything whatever–to help the oppressed.

  3. Thanks Mura. Cummins had the Ethnologue basis in mind when he made the point. Of course there are limitations to any such database.

  4. Hi Kevin,

    My claimed “motivation” was a bit tongue in cheeek. I really wonder why I bother – Mike used to tell me off for wasting time reading this stuff, let alone replying to it.

    I agree with your points – well said!

  5. I’m sure I quoted Sokal somewhere, but I can’t find it, but here’s what he said about doing his famous hoax: “I’m an unabashed Old Leftist who never quite understood how deconstruction was supposed to help the working class. And I’m a stodgy old scientist who believes, naively, that there exists an external world, that there exist objective truths about that world, and that my job is to discover some of them.”

  6. Hi Mura,

    Thanks for this. I presume you’ve read it, have you?

    It deals with K12 education in US (age 5 to 18), it refers to the Common Core State Standards, speifically to “English Language Arts Standards”. Throughout, It refers to “English Learners” (ELs), without – I suppose for idealogical reasons – saying anything about the students’ language backgrounds (home languages, for example), and this makes it difficult for me to understand what they’re proposing with regard to teachers working to achieve “English Language Arts Standards” and teachers of “K–12 English as a Second Language (ESL)”, like Rachel, mentioned in the first line of the article. At least they seem take “English as a second language” as a “legitimate” subject, but they nowhere make it clear,to me, anyway, when they are talking about ESL and when about teaching inside the English Language Arts curriculum. A lot of the article strikes me as vague, the usual strident assertions are made without any empirical evidence to support them.

    Section 7 gives an example of “how teachers can plan CCSS-aligned instruction using the CR (communicative Repertoire)”. Mr. Z’s class is callled “English for English Language Learners”, there are seventeen students from twelve different countries in it, they are in 12th grade, so 17-18 yrs old, and “most students have enough English to write at least a few sentences”, which implies that some of the students don’t. We need to know the syllabus that Mr. Z has to follow, what kind of objectives it has, and how his students will be evaluated. We should also get a clearer exolanation of the authors’ views on the syllabus and evaluation process. Mr Z, anyway, has to teach a syllabus which includes a unit on poetry, and the article shows how he adapts the CCSS.ELA-LITERACY.L.11-12.5 requirements by using CR tenets. The attention given here to exploiting the students’ knowledge of diverse languages is, I think excellent, but there are lots of questions left unanswered, including what view of (second) language learning informs the approach (e.g., how much is language learning developed through implicit learning and what is the role of explicit learning?); what kind of course is Mr. Z’s “English for English Language Learners”? (ESL, or CLIL, or English literacy, for example); what kind of syllabus is used?; and what did the students end up being able to do with English at the end of the course? Typical of the article is the bit below, which I think captures what’s wrong with it: it’s too vague, in its treatment of theories of second language learning, in its discussion of curriculum design, and in its descriptions of alternative teaching practices

    “Throughout the year, as Mr. Z continued to cycle through the 3Ds of the CR approach, students continued to develop their English repertoires as well as to draw on their multiple repertoires—their own languages, additional languages in the classroom, and social media and digital tools—to convey messages for multiple audiences”.

  7. Hi Geoff
    I only glanced at paper, their use of “consciousness raising” reminded me of some corpus linguistic approaches in classrooms; i guess one could chase up the first authors work on “comminicative repertoire”?

  8. it is notable that Garcia (2009) makes this point about Ethnologue;

    so I was curious if Cummins representation of Garcia’s work is fair? for example just after the point Garcia (2009) makes on Ethnologue she writes:
    “Language is truly a social notion that cannot be defined without reference to its speakers and the context in which it is used (Heller, 2007). It is also true, however, that language is a psychological and grammatical notion. For example, the mental grammars of person A and person B may be more similar to the grammars of person A and person C, although whether A and B end up speaking the same language is a sociopolitical decision. It is important then to recognize that, despite the fact that language has a psychological and linguistic component, it is the social context in which it is used, and the wishes and power of its speakers, that determines its role – especially in schools”
    Garcia (2009: 29)

    the above is related to the issue that Cummin’s claims that Garcia doesn’t think languages exist; Garcia & Lei (2014) use a quote from Makoni & Pennycook (2007) –
    “languages do not exist as real entities in the world and neither do they emerge from or represent real environments; they are, by contrast, the inventions of social, cultural and political movements.” (Makoni & PennyCook 2007: 2)
    but the fuller quote continues:
    “On the other hand, we would argue for the very real material effects of linguistic inventions since they influence how languages have been understood, how language policies have been constructed, how education has been pursued, how language tests have been developed and administered, and how people have come to identify with particular labels and at times even to die for them, as the violent nature of ethnic rivalry in Africa, South Asia and elsewhere amply demonstrates. Thus, while the entities around which battles are fought, tests are constructed and language policies are written are inventions, the effects are very real.”(Makoni & PennyCook 2007: 2)

    I wonder if Cummins is playing on a rhetorical device -> they say “languages don’t exist” WTF!?

  9. Hi Mura,

    I think Cummins fairly reports the nonsense García (2009) and García & Lei, 2014) talk about languages. Both repeatedly question the “legitimacy” of referring to English, Spanish and Hindi, for example as distinctly different languages.
    * Language is “a social notion” that can’t be defined, etc., etc.
    * To say that language is a psychological & grammtical notion means nothing without explaining what that means.
    * The “mental grammars” of person A and person B (whatever they turn out to be) “may be more similar to the grammars of person A and person C, although whether A and B end up speaking the same language is a sociopolitical decision”. I.e., who’s speaking what language is a sociopolitical decision!
    * Tests and language policies are written and constructed around “inventions”.

    We’re in Humpty Dumpty land.

  10. re if Cummins is fair – on the Ethnologue point I say he is unfair since Cummins comments “It would also be illegitimate for websites such as Ethnologue ( to refer to and provide information about the 7,106 languages and dialects that humanity has generated.” but Garcia is not saying Ethnologue is illegitmate simply “there is no consensus on the number of languages in the world”.

    as far as i can see only Cummins uses “illegitmate” in the relevant cases?

  11. Pingback: Li Wei on Translanguing | What do you think you're doing?

  12. Pingback: Multilingualism, linguanomics and lingualism | Adaptive Learning in ELT

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s