Dec 10, 2016

One is better than none

One of my students showed her vocabulary (and grammar) notebook to her private tutor, who was surprised at the way new vocabulary was recorded in it. The student then conveyed the tutor's concerns to me, for example, that "pack in" doesn't have to go necessarily with the job (I'd taught the group "she's packed in her job"). She said, "it means 'finish' or 'give up'". I agreed. But where does it get you? If "pack in" can be substituted for "finish" or one of the other alleged synonyms (alleged because no two or more words are ever absolute synonyms - see HERE), can we say "I've packed in my homework"?

It is true that "pack in" collocates with other nouns/noun phrases, but "job" is one of the most frequent collocates. I've consulted two dictionaries - I often look up a word in different online dictionaries to glean its most typical use - and both Oxford Advanced Learners and Macmillan dictionaries give an example with "job", a fact you can't just brush aside when teaching lexis! If two authoritative dictionaries complied by lexicographers using corpora provide similar, if not identical, examples, it surely tells us something about the word usage. What follows are my thoughts on how word usage shapes word meaning and why it's more effective to start with ONE prototypical use.

The linguistic equivalence fallacy

The initial challenge learners face when encountering a new L2 word the first time is to get a grasp of its meaning. One of the easiest ways to 'unlock' the meaning is via L1. Learners tend to mentally translate new items into L1 regardless of whether the teacher uses translation in class or not. L1 mediation is often inevitable and sometimes actually necessary to clarify meaning. Try explaining the word "(to) happen" without translation!

Unfortunately, when mapping a new English word on the existing L1 word or its nearest equivalent, everything that students know about the L1 word - its semantic relations, collocational field, grammatical behaviour - tends to be copied and assigned to the new L2 word. Learners seem to erroneously believe that since both new L2 word and its L1 counterpart refer to the same concept, they will function more or less alike. This assumption, referred to by psycholinguists as semantic equivalence hypothesis (Ijaz 1986, Ringbom 2007), is succinctly summarised by Michael Swan (1997) as follows: "Foreign words look different from mother-tongue words, but work in the same way (semantically and grammatically)".

There are a number of problems with relying on mediation via L1. First, semantic boundaries of seemingly corresponding words in the L1 and L2 do not overlap (as discussed extensively HERE). Second, the corresponding word in the learner's mother tongue may be of low frequency and utility. For example, "burden", a two-star word according to Macmillan Dictionary, which indicates that the word is in the medium-frequency band (5000 most common words in English), is rendered into Russian as бремя (bremya), a rare and even somewhat archaic word. Learners therefore may mentally file the word "burden" under "infrequent and not very useful". The biggest problem, however, with relying on access to L2 words via L1 translation is that it often results in transfer of collocational pateterns from learners' mother tongue to L2.

A lateral road with genuine squares

A student once asked me if she could say a "lateral road" (she must have wanted to say a "bypass"). In the previous lesson, "lateral thinking" had come up and I had encouraged learners to record it as a single item, without breaking it down. I don't know what mental processes she went through to arrive at "lateral road". I do know that learners (especially adults) have a tendency to approach new lexis analytically in that they tease chunks apart and zero in on their components, as opposed to native speakers who are more holistic learners (Wray 2002). I can imagine "lateral" meaning something like "side", "branch" or "offshoot" in other languages, hence my student's conclusion that a "lateral road" would be a possible collocation in English.

Such miscollocations (lexical mismatches) are common among learners at all levels (Laufer &  Waldman 2011) and are usually a result of what used to be referred to as interference (in the Behaviorist tradition), but is now known as transfer; the more neutral, all encompassing term cross-linguistic influence is also used today. Research shows that over 50% mistakes learners make with collocations exhibit signs of transfer from their L1 (Nesselhauf 2003).

While we, teachers, cannot exert much control over learners' internal mental processes, we can do a lot about the way we present and practice new vocabulary. The common practice of clarifying meaning using synonyms should be used with caution. The meaning of "conceal" can be clarified through its near-synonym "hide", but can you *conceal under the table? And "trapped" means to be stuck, but you wouldn't say that your key was *trapped in the door, would you? The other day a student spoke of "a genuine square" while desribing a city. I guess I shouldn't have used "authentic" to explain "genuine" a week earlier; but then, Oxford Learners Dictionaries does the same :)

A "genuine" square in Rome ?
Photo by Tzvi Meller

Does the word 'synonym' have a synonym?

Using synonyms to clarify meaning is problematic also because it gives learners a false sense that any of the two words - the new item and the already familiar synonym - can be substituted for each other in a sentence in any context. I qualified the previous sentence with "in any context" because in some contexts they can indeed be interchangeable - they are synonyms, after all! Like in my example She's packed in her job, "pack in" can be substituted by "quit". Because synonyms only exist in certain contexts, it would be more accurate to talk about synonymous collocations rather than synonymous words, for example:

severe punishment = harsh punishment
severe expression = stern face
severe disability = (very) serious disability

Unfortunately very few coursebooks treat synonymy this way and instead offer exercises where students have to match synonymous words out of context. This is especially common with multi-part verbs:

find out = discover
look into = investigate  

If a new word is clarified by means of a synonym, without any co-text (textual context - see HERE), students might not know how to collocate it; in fact, they might end up not using the new word at all. Several researchers have pointed out that learners tend to avoid synonyms (Liao & Fukuya 2004, Siyanova & Schmitt 2007). Batia Laufer, for example, refers to synonymy  as one of the difficulty-inducing factors in L2 vocabulary acquisition (1997: 172). The reason - the way I see it - is that learners fall back on already known vocabulary. To illustrate, let's take "conceal". If it means the same as "hide", why do I, as a learner, need it if I can get by with "hide"?

One rather than none

How can we make learners incorporate more sophisticated, less common, medium frequency vocabulary into their production and thus push them past the intermediate plateau? Start with one, archetypal use of the word - like I did with "She's packed in her job". Consulting a dictionary or two is a good idea, looking it up in a corpus is even better, but you can also make do with the context the new item first came up in - in a reading text, a podcast, video or any other classroom activity. Review and practice the new item in the original, restricted context until learners commit it to memory and can recall it when needed. Gradually expose them to other uses of the word - the chances are that they will encounter it themselves in other contexts. My students often tell me how they start encountering what at first seemed to be rare, not very useful words everywhere. In fact, there is a name for this process: the Baader-Meinhof phenomenon - look it up! :)

You might say that I'm impeding my students' progress* by deliberately restricting their ability to use a new word. I say it's better to get ONE typical usage right than end up with NONE and have the new word consigned to your passive vocabulary, lying dormant at the back of your mind. Besides, too much information in the early stages of learning may overwhelm the learner and unnecessarily increase the learning burden. Many L2 vocabulary researchers make similar claims suggesting that the depth of knowledge should be increased gradually (Nation 2001; Schmitt 2010); only of course they believe the meaning should be learned first and collocations acquired later from exposure, whereas I think that at least one useful collocation should be provided - as a crutch - from the outset.

Have you tried a similar approach? 
Do your learners struggle to integrate new lexis into their production? 
What do you do about it?
I'd like to hear your thoughts in the comments below.

*Incidentally, I recently taught "impede" initially as impede the progress although, of course, other things can be impeded too, e.g. development, process.


Ijaz, I. H. (1986). Linguistic and cognitive determinants of lexical acquisition in a second language. Language Learning, 36(4): 401-451

Laufer, B. (1997). The lexical plight in second language reading: Words you don’t know, words you think you know, and words you can’t guess. In Coady, J. & T. Huckin (Eds.), Second Language Vocabulary Acquisition (pp 20-34). Cambridge: CUP

Laufer, B. & Waldman, T. (2011). Verb-noun collocations in second language writing: a corpus analysis of learners’ English. Language Learning, 61(2): 647-672

Liao, Y.D. & Fukuya, Y.J. (2004). Avoidance of phrasal verbs: The case of Chinese learners of English. Language Learning 54 (2): 193–226

Nesselhauf, N. (2003). The use of collocations by advanced learners of English and some implications for teaching. Applied Linguistics, 24(2): 223–242

Nation, I.S.P. (2001). Learning vocabulary in another language. Cambridge: CUP

Ringbom, H. (2007). Cross-linguistic similarity in foreign language learning (Vol. 21). Multilingual Matters.

Schmitt, N. (2008). Instructed L2 vocabulary learning. Language Teaching Research 12(3): 329-363

Siyanova, A. & Schmitt, N. (2007). Native and nonnative use of multi-word vs. one-word verbs. IRAL 45: 119-139

Swan, M. (1997). The influence of the mother tongue on second language vocabulary acquisition and use. In McCarthy, M (Ed.) Vocabulary: Description, acquisition and pedagogy (pp 156-180). Cambridge: CUP

Wray, A. (2002). Formulaic language and the lexicon. Cambridge: CUP


  1. Having come across the idea that words sharing patterns can and do share 'aspects' of meaning (Francis & Hunston), I often searched in vain for individual items which shared collocations, alas a search often in vain. Interesting article and cheers for the Baader-Meinhof phenomenon.

    1. Thank you for stopping by and for your comment, Peter!

  2. Great article Leo! Congrats! As a certified language coach I've been implementing some of the lastest discoveries for vocab learning from the neurolinguistic perspective (learning deeply in a more proactive way, using learnt vocabulary in a more personalized way, reuse new words in different contexts and revisiting them from time to time (at least 5 times) it's time consuming and it sometimes gives students the perception that we're not going forward since we're constant revisiting "learnt" vocabulary. However, results have been amazing. As you said, one is better than none. I'm finally seeing some results as opposed to pages and pages of notes that students will never be able to thoroughly assimilate/use.

    1. I see what you mean about students not understanding (at first) the value of revisiting supposedly known vocab. As a wise man said, revision and recycling does take a lot of time, but it's time well spent!

  3. Couldn't agree more Leo, one is definitely better than none, and quality of elaboratin is better than quantity of superficial study! This year my advanced learners have been working systematically with the SkeLL corpus in class precisely to develop their awaremess of mid frequency collocation and the results are quite exciting. At least they are to me :-)

    1. Hi Sharon,
      Would like to hear more Have you written up anything about it over on your blog?

  4. Thanks Leo. I try to get students to think about and record what they'd actual say with new items of vocabulary in addition to nothing down information about the meaning of the item.

    Regarding the question of the lateral road, I wonder if the student was a Spanish speaker translating from the L1. Here in Mexico, 'el lateral' / ' carriles laterales' refer(s) to the lanes that run alongside a high speed main road and are used get on /off the main and to provide access to the buildings, businesses, neighborhoods etc situated next to the main road. Funnily enough, to provide a nice example of your ideas in practice, I can't think of a good translation of carriles laterales that sounds natural in English, but, more importantly, I know exactly what to do when the GPS on my phone tells me to a 'salir a los carriles laterales'.

  5. Thank you for your comment. Good idea to record what students would say.

    Regarding the lateral road, her L1 is Romanian actually (which also belongs to the Romance family), but she hardly uses it. Her L2 is Hebrew; so English is her L3. So it could be the case of translation from L1 (your hypothesis) or it could be the case that she went home, looked up "lateral" in a dictionary, disassembled the chunk as it were, and then tried to build a new utterance in class.

  6. Perhaps the student googled 'lateral road' or looked in a corpus? 'lateral road' appears to be a more technical and possibly more US English alternative for 'side road' or 'branch road' and is certainly valid. Here are some google books results:,cdr:1&num=100&lr=lang_en&gws_rd=cr&ei=psJPWJ3XKsmfgAaLrrLwAw

    I've done this myself when doing learner corpus error annotation - assumed a word or structure was wrong because I personally didn't know or use it. I used to think 'an opportunity *of* doing something' was wrong, and correct it to 'opportunity to do something'.
    In this case, the student wasn't *wrong*.
    I think it's important when asked these 'can I say' questions, to check not against one's own internal 'corpus' but against what's out there outside of our ideolect or our world knowledge.

    1. Hi, Diane

      I very much doubt she looked it up in a corpus (I wish she had!) but I did and couldn't find ROAD in the list of collocates of "lateral" in COCA, even after increasing the no. of hits to 500 (under Options). Now I know, of course, that corpus doesn't provide negative evidence, i.e. it doesn't mean that "lateral+road" is not possible.

      I followed your link, which I repost here for other readers to see:,cdr:1&num=100&lr=lang_en&gws_rd=cr&ei=psJPWJ3XKsmfgAaLrrLwAw
      (somehow commmenters's links do not come out as hyperlinks and are not clickable)

      Most of the uses of "lateral road" in Google Book search are in law books dating from the 19th century - hardly the use she was going for. So I don't think my guidance or "correction" was entirely wrong. Or do you think it was?

      Thank you for joining the discussion and reminding me that I've been meaning to blog about idiolect.


    2. Hi Leo,
      If you sort the result in the url by date, rather than relevance, you get the most recent books first. These include recent fiction, history etc books. It's still the case (as I said myself) that it seems to be used a lot in legal contexts, but definitely also other more mainstream genres. I find two cites in COCA and two in the NOW corpus (i.e. uses this year). 'Lateral roads' also yields results and has the handy side-effect of excluding the proper noun use for the name of an actual road in Bhutan.
      It is, of course, not a strong collocate of 'lateral'. If you had searched for collocates of 'road' instead of collocates of 'lateral', I think you would have got a more helpful result for your student, as 'side' is #3 with 2,738 uses, and given that a standard dictionary definition of the first sense of 'lateral' is, basically, 'side', you could have said, well, I've never heard of a 'lateral road', though it is used in quite a lot of contexts and it's quite clear what you mean to anyone who knows the core sense of lateral (i.e. side), I think more people will understand what you mean if you say 'side road', because that's more common.
      You ask whether I think you were wrong to tell her she was wrong. Whether you were right or wrong was never really my point. I just wanted to point out (as I point out to myself often) that the limits of one's own lexical arsenal are not the limits of the language's. But yes, since you said this:
      "I can imagine "lateral" meaning something like "side", "branch" or "offshoot" in other languages, hence my student's conclusion that a "lateral road" would be a possible collocation in English."
      Since 'lateral' means something like 'side', 'branch' in English too, you were wrong to say it doesn't work in English. There was no need for speculation about transfer etc. because 'lateral' means what she thought it meant and most people would understand what she meant, even if they'd never encountered the term.
      Really looking forward to your post on ideolect, Leo! I always find your writing engaging. And thanks for this post, which was very thoght-provoking!

  7. Great post Leo. L1 transference is a massive problem with teachers constantly having to tell students: "because that'she the way it IS in English so you must learn the words patterns, how it behaves,collocates and colligates. I've started highlighting and emphasising how colligation and cotext are essential to learning new lexis which means boarding words and collocations with their grammatical patterns and cotext (dialogues,or follow-on sentences). More is more so it's a win win situation. As teachers we need to go beyond the cursory collocation or chuck and start becoming attentive to how certain lexis colligates and generates cotext around it. Teaching collocations and chunks isn'enough. Love your posts, keep them coming

    1. Thank you for your comment.

      I agree that more is more and providing whole sentences with follow-on sentences and thus exposing Ss to colligation /grammatical patterns etc. is much better than just collocation. But when there is a time pressure, a cursory collocation/chunk is better than a cursory single word gloss, don't you think?

      Glad you're enjoying my posts!

  8. This is an excellent post and provides solutions to vocabulary issues I have been struggling with lately. I plan to keep this stored as I work on revamping my vocabulary instruction as I feel you have provided a strong and well-researched guide for vocabulary acquisition. Thank you!

    1. Thank you, Bret. Glad you're 'bookmarking' my post for future reference.

  9. Great article, Leo! This topic brings to my mind an activity suggested to me by John Sivell of Brock University's Department of Applied Linguistics. Students choose a few of the new words from the text being studied and find as many synonyms as they can for each. Then they ask themselves and each other whether each synonym can or cannot substitute for the original. I have done this with my multilevel seniors class, and it was provoked a lot of good discussion re collocation, nuance, register, etc.

    1. Thank you for your comment, Kelly (and sorry for my belated response).
      Sounds like a great activity. I don't know why though it has to be NEW words - I think known vocabulary would do just as well. I'm going to try it!


Related Posts Plugin for WordPress, Blogger...