A Few Remarks in Defense of the Markedness Model (MM)

最近去慧大王的課上導讀一篇關於 Markedness Model (MM) 的文章，聽到有些評論，在此回應一下。由於文章用英文寫的，我就用英文來說明。倒數第二段是我覺得最重要的想法。

About a week ago I heard some critical comments on the Markedness Model (Myers-Scotton, 1998) that I think miss the mark. While the comments do point out a few problems of the model, these problems are not unique to MM but are faced by science in general. Let me elaborate.

The first criticism is about the narrowness of MM's explanatory scope. It goes like this: Some students in an English classroom would sometimes switch to their native language to talk with each other, even though the teacher has established an English-only rule for the class. The students use a marked code (their native tongue) for the English-only setting, but they have no intention to change the rights and obligations associated with this setting. How can MM explain this?

The answer is quite simple. While MM proposes a few explanations for the various degrees of markedness in people's linguistic use, it makes no claim that it has exhausted all possible explanations. The students in the above scenario may simply feel that they can communicate better with their peers in the mother tongue, and this reason does not falsify MM because, again, MM does not claim that negotiation of rights and obligations is the only reason people use marked code.

Incidentally, if the students in the scenario above use the marked code (i.e., their L1) because their English (i.e., the unmarked code for the setting) is not proficient enough for communication purposes, MM can, in fact, account for this. In this case, the unmarked code, English, is never in the students' opportunity set to begin with because it has been filtered out by the model's first filter (see Ch. 2 of Myers-Scotton 1998 for the three filters). If English is not really in the opportunity set, there is no rational choice to speak of.

The second criticism would turn my reply above into an admission of MM's flaw. That is, by saying MM does not mention other potential reasons for using marked and unmarked code, I have admitted it is too "narrow and outdated" in its explanatory power. This type of criticism would point to newer accounts (usually Translingualism) as its alternatives or even replacements.

To that, I would reply by saying that the criticism sets an unreasonably high bar not just for MM but for scientific models in general. Newton's first law of motion--the law of inertia--makes predictions about a very narrowly defined situation, in which a body is not acted upon by a force, such as air friction. The law does not apply to most of what we see every day on Earth, where moving bodies are constantly faced by air friction. Yet, I haven't heard criticisms of the law's "narrow explanatory scope." I see no reason why this kind of criticism should be leveled against models like MM. Keep in mind that interactions between forces in the physical world (think of friction, gravity, air density, wind, etc.) are just as complex as (if not more so than) those between cognitive modules in the mind and between agents in a social setting. There is no reason why physics should have the license to be "narrow," while cognitive and social sciences should not.

Furthermore, to the extent that I can tell, Translingualism is not a model for explaining the various degrees of markedness in people's linguistic choices. As far as I know, there is nothing in Translingualism that falsifies what MM claims, and therefore it cannot "replace" MM. (Just as Translingualism cannot fully replace L2 writing, in my opinion.) At best, it has the potential to "integrate" MM and become a more explanatory model by drawing a connection between markedness and something other than negotiation of rights and obligations. I do not see this happening in the field of Translingualism right now, and I invite critics of MM to point out any effort made on the Translingualists' part to integrate MM.

In fact, I would argue that what the Translingualists attempt to achieve with their new pedagogical approach is exactly what MM predicts. Many would agree that Translingualism is a challenge to the status quo, a way for ESL students to find their agency in their writing by using English and other semiotic resources in a manner different from what the "standard" requires (which roughly equals Anglo-American academic writing conventions). Translingualism is, in a way, telling writers to feel confident in their use of what has been traditionally seen as "marked" (e.g., Indian or Singaporean English) in the academic setting, where American and British English are the unmarked. By doing so, the writers are engaged in a negotiation of rights and obligations with gate keepers (e.g., whether a classroom essay peppered with Chinglish expressions should be given an A+ by a teacher, or whether an EFL researcher's manuscript should be accepted into a conference or journal by an editor). The way I see it, Translingualism does not replace MM; it is actually what MM predicts would happen in the field of academic writing, given the field's long history of power imbalance between those who have a good command of prestigious varieties of English and those who don't.

Finally, a reply to the criticism with respect to MM's lack of predictive power: To be sure, even Myers-Scotton admits that MM does not have much predictive power in any broad sense. I would say, however, that the model can be considered predictive in the narrow sense that upon close and honest introspection (which is a kind of observation, by the way), we would, from time to time, find that our linguistic choices are driven by a desire to negotiate rights and obligations. In human sciences, this is sometimes the best we can get: a prediction of something happening with a certain frequency that isn't 100%. (Note that in medical sciences, this is sometimes true as well.) We can only hope that at some point in the future, we will have better tools or methods at our disposal to test these models with higher precision.

Reference:

Myers-Scotton, Carol. (1998). A theoretical introduction to the markedness model. In Codes and consequences: Choosing linguistic varieties, ed. Carol Myers-Scotton. Oxford University Press.