An Interview with Dr. Nan Jiang

Page No.: 
Daniel Dunkley

Welcome to the January/February edition of TLT Interviews! In the first issue of 2019 we are delighted to share with you two fascinating conversations, one on cross-linguistic influence and the other on curriculum methodology. Our first interview is with Dr. Nan Jiang. Dr. Jiang is an associate professor in the School of Languages, Literatures, and Cultures at the University of Maryland where he teaches second language acquisition. His research field is the psycholinguistics of adult second language acquisition, and his research papers have appeared in journals such as Journal of Neurolinguistics and Applied Psycholinguistics. He was interviewed by Daniel Dunkley, an English lecturer at Aichi Gakuin University, Nagoya.

An Interview with Dr. Nan Jiang

Daniel Dunkley: Thank you for this interview, Dr Jiang. To begin, what do you mean by cross-linguistic influence?

Dr. Nan Jiang: This is an old topic that has been around for at least four decades. When people started looking at the language acquisition issue, they realized that the learner’s L1 plays a very significant role. It started in the theory called contrastive analysis in the 1950s. Now, the study of cross-linguistic influence has expanded tremendously. In the early days, we analyzed the learner’s errors to see whether you can trace these errors back to the learner’s L1. Later, we looked at language processing data in different aspects of language: phonology, vocabulary and syntax.

How about your own cross-linguistic influence research?

My own work has been in the area of semantics and morpho-syntax. For instance, in the area of semantics, each language has its own unique semantic structure, both at the lexical level and at the language level as a whole. For instance, in Japanese you have the concept of yohkai, which is a monster-like creature which can be a good yohkai or a bad yohkai. We have the same concept in Chinese: yohwai. In English, we don’t have such a concept, so in order to learn the word yohkai for English native speakers you have to learn this new concept. In this regard, you have a difference between what has been lexicalized in one language and in another. That’s an example of a semantic difficulty for English L1 students of Japanese or Chinese.

Let’s consider a similar problem for Chinese learners of English. In English, you have two words: criterion and standard, but in Chinese we have one word, byojun (hyojun in Japanese). The semantic distinction between the two words exists in English but is absent in Chinese. So, the L1 Chinese student of English has to learn a new distinction, which can be very difficult. That’s one area where semantic transfer occurs a lot. There are many examples from my own experience: even though in theory I know the difference between cap and hat, I sometimes use these words interchangeably. This is because this semantic distinction is not rooted in my mind.

That deals with semantics. How about morpho-syntax?

In terms of morpho-syntax, Chinese, like Japanese, doesn’t mark plural obligatorily, so we have a huge problem with plural marking. I have done some research showing, firstly, that it is an authentic difficulty among Chinese native speakers. It’s not that they have already developed English-specific plural marking, but for some reason, they forget to use it—that’s the processing explanation of this difficulty. My view is that this difficulty is at the representational level; they haven’t developed a native-like knowledge about plural marking. They have the explicit knowledge for sure because that’s what’s taught in the classroom, but they are not able to use such knowledge in spontaneous communication. My research also shows that plural marking errors have a lot to do with the first language. For instance, Russian native speakers don’t seem to have that much difficulty with plural marking in English, but Chinese and Japanese native speakers do. To complete my summary, we must mention phonology. After a certain age, you begin to have an accent when speaking a second language. In short, these three areas- semantics, morpho-syntax and phonology, have constantly shown a cross-linguistic influence from L1 to L2.

Is there any hierarchy of difficulty between these three areas?

Phonology is definitely the most difficult. The foreign accent is common after a certain age. In fact, very few people can develop native-like pronunciation after puberty. Some researchers have abandoned the term critical period for the term sensitive period, for the reason that the loss of a certain ability is gradual rather than sudden.

Pronunciation seems to be a hopeless case, but is it possible for adults to overcome L1 influence in grammar and vocabulary as a result of study?

As far as plural marking is concerned, I haven’t seen many Chinese native speakers who have developed native-like competence in this area after living in an English environment. In my own case, I’ve lived in America for 25 years, and I still have difficulty with plural marking in spontaneous speaking. On the other hand, in writing, it’s less of a problem because I can always correct myself. As for vocabulary, learning semantic distinctions is slow, but with time, one may approach native-like competence.

You’ve talked about Japanese and Chinese languages, which lack many features of English. In some respects, they are simpler than English. What about English students with grammatically more complex L1s, such as Russian or German?

If something is marked in their native language—for example German marks the plural or marks the past tense—then that shouldn’t be a huge problem. On the other hand, when we study English, L1 learners of French or Spanish have great problems mastering grammatical gender marking.

How about the articles in English?

We published a paper showing that if your language doesn’t have an article system, it’s extremely difficult to develop a native-like use of the articles.

In teaching, do we put more emphasis on teaching articles or just give up?

It’s a tough question. Articles are very complicated in the sense that they are associated with different meanings and different uses. Some uses are easier to learn than others. For instance, the zero article is difficult for me—that is when no article is necessary. We can’t talk about the learning or teaching of articles as if it is a monolithic concept. There are different uses of articles. Some uses can be taught explicitly, and it may be effective. For example, with the use of sports and musical instruments, we play the piano, but we play baseball. But even native speakers are not completely consistent in their use of articles. For example, contrary to the apparently simple rule I’ve just given, we often hear people say, “I play guitar.” In short, there are certain rules that are useful, and others which are less helpful.

How do you do cross-linguistic research?

At first, people used to collect lexical errors as evidence for language transfer. Now, methods have changed. My research is more lab-based. At the lexical level, one topic I’m looking into now is whether adult non-native speakers (NNS) can develop new semantic distinctions. One study I did was to compare NNS’s reaction time on pairs of English words that share the same L1 translation and pairs of English words that do not. As I said earlier, standard and criterion are always translated into the same Chinese word, so a semantic distinction is not made. By contrast, an example of a different translation pair is interfere and interrupt. These are related in meaning, but they have different Chinese translations, so a semantic distinction is made in both languages. What I asked people to do was to decide whether two English words are related in meaning. They press the yes button if they think they are related and the no button if they think they are not related. What is interesting is their performance on the related items. The related items are divided into similar translation pairs and different translation pairs. By looking at the reaction times between the two types of pairs, we can get a sense of whether these pairs of English words are still mapped to the same Chinese concepts. If they are, then the Chinese people should respond to the same translation pairs faster than different translation pairs. English native speakers should respond to these two sets in the same amount of time, showing that they are very similar in terms of semantic relatedness.

Finally, what should teachers in Japan bear in mind about cross-linguistic differences?

The first thing is to know that there are cross-linguistic differences. Many teachers know that. The second thing is to understand that such differences cause problems for learners. And if they are aware of such differences, they can provide explanations when necessary. For example, the Chinese word “sofa” comes from sofa in English. But we can use the word safa in Chinese to mean a padded chair. So, explicit explanation might help. Another approach would be to provide targeted input so that the learner may come to their own understanding about differences between words. For example, the word meeting is different in English and Chinese. A meeting in Chinese always has more than two people, whereas two people can only have a talk, not a meeting, while in English you can have a meeting with two people. To summarize, my recommendation is that you provide targeted input and explicit explanation.

Thank you, Dr Jiang, for your lucid explanation of this topic.


Jiang, N. (2002). Form-meaning mapping in vocabulary acquisition in a second language. Studies in Second Language Acquisition, 24, 617-637.

Jiang, N. (2016). Conducting Reaction Time Research in Second Language Studies. London, Routledge.