According to my Catalan teacher, Dr Josep Sobrer of
Teaching
vocabulary in semantic sets
Perhaps the most striking new idea that I have learned from this class is that “Words that are formally similar to each other, or that belong to the same lexical set, or which are near synonyms, opposites, or free associates should not be learned together.” (Nation, 2001). This is very much in conflict with the fact that “almost all second language textbooks arrange vocabulary in semantic sets.” (Folse, 2004).
According to Tinkham (1997), there is “little or no empirical justification” to support teaching in lexical sets, such as items that would “fall under a common superordinate,” which he calls semantic clusters. In fact, he claims that “a large body of evidence...suggests just the opposite,” referring to “interference theory” from the field of psychology. He claims that “interference theory,” supported by “hundreds of studies” from psychology, would predict that learning semantic clusters would impede learning as “similarity between information...increases the difficulty.” He hypothesizes, though, that
learning words of different parts of speech that might participate in certain “frames” or “schema,” such as frog, pond, hop, swim, green, and slippery, which he calls thematic clusters, might be beneficial.
He conducted a study on 48 subjects, all native speakers of English at a large mid-western American university. Students were tested on their recognition and recall of artificial words in both oral and written mode. He measured the numbers of trials (attempts) that each student had to make in order to learn the words. He compares the learning of semantic clusters, thematic clusters, and unrelated sets. He found a “clear indication that new L2 vocabulary items arranged in semantic clusters are learnt with more difficulty than new vocabulary items learnt in unrelated sets.” He reports, “Of the 96 comparisons contrasting the trials needed to learn a semantic cluster to the number needed to learn an opposing unrelated set, the semantic cluster was learnt in more trials than the opposing unrelated set in 80 of the 96 comparisons.” He concludes, “semantic clustering does indeed serve as a hindrance.”
Results for the thematic clustering were “mixed, but generally positive.” He reports, “Of the 96 comparisons contrasting the trials needed to learn a thematic cluster to the number needed to learn an unrelated set, the thematic cluster was learnt in fewer trials than the unrelated set in 47 of the comparisons, in more trials in 20 of the comparisons, and in the same number of trials in 29 comparisons.” Thus, “thematic clustering was a benefit to learning only half the time and actually a detriment about one-fifth of the time.” When students were asked which set was easiest to learn, the “vast majority” identified the thematic cluster as the easiest, and none chose the semantic cluster.
Waring (1997) replicated Tinkham’s study with similar results. Waring also reports that mixing the vocabulary, i.e. producing an “incorrect word taken from the same set,” occurred 25% of the time with words in a related set, but only 5% of the time in an unrelated set. He concludes that although grouping words in semantic sets may be acceptable for reviewing words, presenting new words in semantic sets should be avoided.
Before reading this research, I could never have imagined the detrimental effects of teaching vocabulary in semantic sets, yet I have always lamented the fact that textbook authors seem to place a higher priority on adhering to a semantic set when creating readings and exercises than on creating readings and exercises that will interest the students. One can only hope that this research will one day bring an end to the readings and exercises contrived to adhere to a semantic set. The findings about teaching vocabulary in thematic groups rather than semantic groups comes as good news, as thematic groups should occur naturally with little or no effort on the part of the materials creator. Also the fact that “thematic clustering was a benefit to learning only half the time” (Tinkham, 1997), will hopefully prevent materials writers from simply replacing the current absolute insistence on semantic clustering with a “new” absolute insistence on thematic clustering.
Discourse
length context
According to Omaggio (1986), language can best be learned in “communicative contexts” not “random lists or disconnected sentences.” Omaggio feels that students need discourse-length context in order to develop discourse competence. Although Omaggio recommends learning in context from the beginning of instruction, not everyone is so insistent on discourse-length context from the beginning. According to VanPatten (1996) “early activities should involve sentence level input” as sentence level input gives learners “ a better chance at focusing on and detecting the targeted form in sentences...Connected discourse...should be reserved for later...[as it] may hinder learners’ initial processing of the targeted form because of their limited capacity to process incoming data.” VanPatten (1999) also states, “the shorter [the sentences] the better.” In fact, Vanpatten (1996) recommends an exercise in which sentences are purposely taken out of context:
Listen as your instructor reads a sentence. First, decide whether the sentence is about Bill Clinton or Hillary Rodham Clinton. Then, decide whether you agree with the statement or not.
(Instructor reads: 1. Es dinámica.; 2. Es agresivo.; 3. Es egocéntrica.; 4. Es honesta.; 5. Es respetado.; etc.)
In normal discourse, the -o/-a gender alternation is almost always redundant in Spanish (it supplies no additional meaning), and therefore the student does not have to attend to it. By removing the sentences from context, the student has to attend to the endings to complete the task.
Prince (1996) performed a study on
48 EFL students in
|
|
|
translation learning |
context learning |
|
weak learners |
translation recall |
85.81 |
44.52 |
|
|
context recall |
32.37 |
32.22 |
|
advanced learners |
translation recall |
74.54 |
56.84 |
|
|
context recall |
74.03 |
63.92 |
Prince concludes that “effective learning of words requires a stage in which the word is in fact isolated from its context.”
Laufer and Shmueli (1997) tested students’ vocabulary learning under four different conditions: 1) isolated words on a list, 2) words in meaningful yet disconnected sentences, 3) words in full text context and 4) words in “elaborated” full text context, i.e. “supplemented by clarifying phrases.” In all conditions, students had glosses of the target words in either L1 (Hebrew) or L2 (English), so there was no need to guess the meaning of the target words by context. The most effective method for “long-term retention” was the disconnected sentence presentation, followed closely by the word list. The least effective method for either short-term or long-term retention was the unaltered text. Furthermore, the list and disconnected sentence methods only took 25 minutes, whereas the text methods took 70 minutes. Laufer and Shmueli feel that minimal context encourages “noticing” and that their study shows “superiority of focus over context.” In fact, it has been shown that word cards, perhaps the ultimate decontextualization of vocabulary, can be a very effective way to learn. According to Waring (2001) “Word card learning can result in huge vocabulary gains.” Nation (2003) also endorses word cards, “...the direct learning of L2 vocabulary using word cards with their L1 translations is a very effective method of learning.”
I have always introduced new vocabulary and grammatical structures to students by creating comical sentences containing the targeted items. These sentences were almost always disconnected, as opposed to being in discourse-length context, due to the fact that it would be very time consuming for me to attempt to write humorous/interesting discourse-length paragraphs containing the target items without introducing vocabulary and structures that were not the current focus of learning. I have always chosen to put the priority on humor/interest rather than on discourse-length context. I had always felt a bit “unprofessional” about this before availing myself of this research, and I am very glad to know of it.
Authentic
materials
According to Omaggio (1986), “Authentic language should be used in instruction wherever and whenever possible.” Nevertheless, she wisely recommends “simplified authentic materials with Novice- and Intermediate-level students” also adding “the task...must be geared to the [students’] proficiency level.” Using authentic materials with beginners conflicts with Nation’s (2001) recommendation that readings should contain 95-99% known words to develop fluent reading skills. According to Nation, “Simplification is an important tool...without simplification, the strands of meaning-focused input...become impossible.” Certainly reading is important, as Huang and van Naerssen (1987) report “reading practice...as the most significant predictor of oral proficiency when examined along with speaking and listening practice.” Pichette (2005), however, finds that extensive reading is not particularly beneficial for lower-level students. Although Pichette performed an excellent study and validated a common sense idea, he did not distinguish between pedagogically prepared readings and non-pedagogically prepared readings. I feel that beginners need simplified readings. Pigada and Schmitt (2006) also recommend simplified readings as more appropriate for vocabulary learning. I think that the simplified cultural readings that appeared in the beginning Spanish, French, German, and Portuguese books that I used as a student were an excellent source of input and a very enjoyable way of learning about the target cultures.
Pictorial
representations
According to a study performed on
native English speakers learning Spanish by Lado,
“...the students’ performance was highest when both pictorial and written annotations were available and lowest when neither type of annotation was available. The students who had both written and pictorial annotations available had a higher performance than the students who had only pictorial annotations available in the immediate vocabulary post-test, but they did not differ in their performance in the delayed vocabulary post-test...the difference in performance between the written annotations [only] group and the [no annotations] group...was statistically significant in the immediate test but not in the delayed test.”
Jones and Plass, thus, conclude that the benefits of pictorial annotations were more persistent than those of written annotations. Although I have not seen a textbook with pictorial representations of all or almost all vocabulary since Usted y yo, the text I used in high school published in 1975, I hope that learning by pictures makes a “come back” as it has been proven effective. Hopefully textbook publishers will one day supply students with conventional and/or electronic “word cards” (via the web) with pictorial representations to accompany their books.
In this class I have learned that there is always “give and take.” Authentic materials can be exciting and useful, but they can also overwhelm beginners. Whenever we emphasize the current chapter’s vocabulary, we are necessarily avoiding recycling. Contextualized readings can be wonderful; I would never trade the simplified readings in the beginning Spanish, French, German, or Portuguese books that I used as a student for a bilingual word list, yet lengthy context can subtract from focus. Students need a “balanced diet,” and this conflicts with our profession’s “absolutes.” Anytime we tend to extremes such as “only authentic readings” or “only contextualized exercises” it comes at a cost. It is ironic that this profession has so many absolute insistences because, of all people, it is the language teacher that should know that there is always “an exception to the rule.”
I envision the perfect textbook chapter as having short simplified readings which use the current vocabulary and also recycle past vocabulary. There would be no effort to follow semantic clusters. These readings would be followed by a “consolidation task” such as a cloze exercise to review newly met vocabulary (Laufer and Shmueli, 1997). The vocabulary lists would have L1 translations and a pictorial representation whenever possible. In addition to L1 translations and pictorial representations, each vocabulary item would have an example sentence to help students focus on it and also to supply important collocation information. These sentences would be short, as according to Folse (2004), “Longer context do not promote noticing.” There would be copious communicative exercises such as personal questions, ranking and sorting activities, and other exercises that would allow students to express their personal opinions. There would be many exercises on the sentence level in which students would mark sentences as true/false, logical/illogical, lucky/unlucky, etc. Hopefully, there would be multiple opportunities to make “retrievals” of each word, as the number of retrievals is so important (Folse, 2004). Students would also be required to learn some vocabulary of their own choosing (Schmitt, 2000).
In conclusion, I would like to add that I have learned something in French 692 that goes beyond teaching vocabulary or even teaching language in general --- I have learned to accept nothing without empirical evidence. Intuition may lead us to ideas about language teaching, yet our intuitions should always be checked empirically.
REFERENCES
Folse, K.
(2004). Vocabulary myths: Applying
second language research to classroom teaching.
Huang, X., & van Naerssen, M. (1987). Learning strategies for oral communication. Applied Linguistics, 8, 287-307.
Jones, L. & Plass, J. (2002). Supporting listening comprehension and vocabulary acquisition in French with multimedia annotations. Modern Language Journal 86, 546-561.
Lado, R.,
Laufer, B. and Shmueli, K. (1997). Memorizing new words: Does teaching have anything to do with it? RELC Journal 28, 89-108.
Nation, P.
(2001). Learning
vocabulary in another language.
Nation, P. (2003). The role of the first
language in foreign language learning. The Asian
EFL Journal, 5. http://www.asian-efl-journal.com/june_2003_PN.php
Omaggio, A. (1986). Teaching language in context:
Proficiency-oriented instruction.
Pichette,
F. (2005). Time Spent on
in Second Language Learning. Canadian Modern Language Review, 62, 243-262.
Pigada, M.,
& Schmitt, N. (2006). Vocabulary
acquisition from extensive reading: A case study. Reading in a
Foreign Language, 18(1).
Prince, P. (1996). Second language vocabulary learning: The role of context versus translations as a function of proficiency. The Modern Language Journal, 80, 478-493.
Schmitt, N.
(2000). Vocabulary
in Language Teaching.
Terrell, T. (1977). A natural approach to the acquisition and learning of a language, Modern Language Journal, 61, 325-336.
Tinkham, T. (1997). The effects of semantic and thematic clustering on the learning of second language vocabulary. Second Language Research 13, 138–163.
VanPatten,
B. (1996). Input processing and grammar
instruction in second language acquisition.
VanPatten,
B. (1999). Processing
instruction as form-meaning connections: Issues in theory and research. In Lee, J.F. & A. Valdman (Eds). Meaning and Form: Multiple Perspectives (43-68).
Waring, R. (1997). The negative effects of learning words
in semantic sets: a replication. System 25, 261–274.
Waring, R. (2001, March 13). A Study of Receptive and Productive Vocabulary Learning from Word Cards. Retrieved November 26, 2006, from http://juppiter.fltr.ucl.ac.be/FLTR/GERM/ETAN/bibs/vocab/wordcard.html