Input in Language Instruction
Input, as noted by Rost (1990), “is what learners are provided, have access to, or are expected to have available (including prior knowledge) as they proceed in a learning activity”. Input is often associated with learning materials and, according to Rost (1990), it includes materials and language data that the learners are to attend to or manipulate during the task.
Selecting input for language instruction requires consideration of the cultural aspects of the content, length of the extract, abstractness of the content, number of information points, and level of linguistic difficulty (Rost, 1990). However, as observed by this author, none of these considerations alone can predict the difficulties the language learner will experience when encountered with the text, since text difficulty is also related to learners’ motivation and interest in the topic. For example, texts that are interesting or vivid may be easy to understand even if they contain complex syntax and low frequency, technical vocabulary (Rost, 1990).
This concept is in-sync with the Input Hypothesis (Krashen, 1985), which states that learners acquire language when they are exposed to comprehensible input containing linguistic forms that are slightly more advanced than learners’ current language system. However, as noted by Krashen (2003), in order to acquire language we need to understand the input: “we acquire language and develop literacy when we understand messages, that is, when we understand what we hear and what we read, when we receive ‘comprehensive input’”.
An additional consideration in selecting input for language instruction is the nature and source of the input. Bacon (1992) pointed out the importance of second language methodologists (Bacon, 1987, 1989; Rogers & Medley, 1988, cited in Bacon, 1992) who investigated the “message” in input and highlighted the need to expose learners to natural and authentic language. This kind of input, called authentic input, provides the learner with both linguistic and cultural information that may not be available in pedagogical texts (Bacon, 1992, p. 398).
An authentic text, as defined by Gilmore (2007), “is a stretch of real language, produced by a real speaker or writer for a real audience and designed to convey a real message of some sort” Consequently, most everyday items in the target language qualify as authentic materials (e.g., menus, brochures, receipts, television programs, newspapers, radio broadcasts, music, literature, among other sources). Authentic materials refer to materials that were not created or edited expressly for language learners (Nunan & Miller, 1995). Thus, it is possible to discern whether a text is authentic or not by referring to the source of the discourse and the context of its production (Gilmore, 2007).
Classroom materials, as noted by Clarke and Silverstein (1977), should resemble the “real world” as closely as possible. According to these authors, “since language is a tool of communication, methods and materials should concentrate on the message and not on the medium” (Clarke & Silverstein, 1977, p. 51). Many methodologists advocate the use of authentic texts as a way of incorporating the advantages of natural acquisition into the formality of classroom learning (Bacon, 1992).
Bacon (1992) noted that although authentic texts are usually not created with the purpose of language instruction, and typically learners do not have an opportunity to interact with the author of the input, “yet these texts provide a model that is more life-like and potentially more interesting than typical pedagogical text” (Bacon, 1992, p. 399).
Conversely, according to Richards (2006), using appropriate authentic materials or authentic source materials is not always realistic. Although finding authentic source materials, especially written texts, is rather easy and likely to be more motivating than author-written texts, it is still necessary to remove low-frequency lexical items and obscure syntax and to accommodate the length and format of the text to the requirements of the lesson.
Finding authentic source materials with the appropriate level of difficulty for the target learners is rather difficult. Authentic source written texts, for example, can be easily found in magazines or the Internet; however, these texts may be written for a specific audience or may not be age appropriate (Richards, 2006).
Moreover, as noted by Richards (2006), written authentic texts present a big challenge even for college-age language learners due to the fact that real world readers are assumed to have a high level of reading ability and substantial word recognition.
Thus, these texts need certain level of adaptation before they can become useful for language instruction (Richards, 2006). Furthermore, authentic texts (i.e., texts originally spoken or written by and for native speakers and not intended for language teaching), except when used at very advanced levels, impede learning by confronting learners with large amount of unknown material (e.g., new vocabulary and complex grammar) without compensatory devices to facilitate comprehension (Long, 2007).
As noted by this author, written or oral authentic texts present too dense a linguistic target due to the lack of redundancy.
Using Authentic Materials for Listening Instruction
The ultimate goal for listening instruction is to help second language listeners understand the target language in everyday situations (Vandergrift, 2007). According to this author, “authentic listening materials are best suited to achieve this goal because they reflect real life listening, they are relevant to the learners’ life, and they allow for exposure to different varieties of language” (Vandergrift, 2007, p. 200).
When language learners are taught how to listen without the threat of a test, they “find it motivating to learn to understand authentic texts since this practice can help them access similar texts in real life listening” (Vandergrift, 2007, p. 200).
However, there are issues of concern regarding the use of authentic texts for language instruction. One issue is that although authentic texts could bring real life situations and real authentic oral interaction to the classroom, caution needs to be taken in selecting instructional materials since some authentic source texts have no pedagogical value to the language learner (Richards, 2006).
According to Richards (2006), especially in the case of speaking materials, there are other concerning issues. Texts need to meet several design criteria (e.g., sentence length, exchange of conversation length, grammar, etc.) if they are intended to present new language, model speaking tasks, or provide content to initiate discussion. Chunks of authentic discourse would not meet these criteria and will have no pedagogical value for language instruction.
As emphasized by Brown and Yule (1983, cited in Richards, 2006), an informal conversation in the real world serves the main purpose of maintaining social interactions. Consequently, the main reason of “chatting” is to be nice to the other person and not to convey information, and thus these texts are of little relevance to anyone else.
An additional concern in selecting input for listening in the language classroom is not only the nature and source of input, but also the purpose of the audience for whom the input is intended (Rost, 1990).
The author emphasizes Widdowson’s (1979) differentiation between the concerns of the text itself and the concerns of learner’s use of the texts. As noted by Rost (1990), “genuineness is a characteristic of the text itself and is an absolute quality. Authenticity is a characteristic of the relationship between the text and the listener or reader and has to do with appropriate response” (p. 160). Rost (1990) indicated that while many language educators advocate the use of prerecorded texts of native speaker conversations because of the genuineness they bring to the classroom, others (e.g., Candlin & Edelhoff, 1982, cited in Rost, 1990) argue that genuine texts do not necessarily lead to authenticity of the purpose for the learner.
Teachers aiming to create real-life conditions of listening in the classroom may not be able to do so using genuine texts (Rost, 1990).
Therefore, some educators support the use of authentic texts only in situations when it is needed, for example, to show dialectal differences or features of settings in particular locales (Ur, 1984, cited in Rost, 1990).
While examples of authentic source listening materials are abundant (e.g., radio broadcast, television announcements, etc.), they are impractical due “to logistical problems involved in recording genuine interactions, and copyright and ethnical issues that arise when one wants to use data from those sources” (Richards, 2006, p. 21).
Additionally, these texts require substantial modification in order to be adapted for language instruction.
In the early years of research on this topic, modifications of input were considered to be changes in linguistic form (i.e., surface syntax, lexis, and phonology) or modifications of interactions involving features of conversation or discourse function (Parker & Chaudron, 1987). As noted by these authors, features of linguistic form modified to less complex ones included shorter utterances, and less complex syntax and vocabulary. Modifications of interaction included clarification requests, comprehension checks, confirmation checks, and completion and repetition of others’ utterances.
Parker and Chaudron (1987) later introduced a third type of input modification – elaboration – which they wanted to distinguish from simplification and modifications involving negotiation of meaning.
Elaboration, or elaborative modifications as defined by Parker and Chaudron (1987), refers to the addition of repetitions or redundancy and clearer signaling of the thematic structure of the communication.
Simplification, as defined by Honeyfield (1977), Involves limiting syntax and vocabulary through de-transformation and paraphrasing. These processes reduce information density, and also disrupt the normal system of information distribution (since low frequency items are not used). Further, the highly restricted syntax that is often used may be inadequate for a given information load, and so may reduce cohesion and readability.
According to Parker and Chaudron (1987), typical features of linguistic simplification for instructional context include the use of shorter utterances, simpler syntax, simpler, deletion of sentence elements or morphological inflections, and preference for canonical word order.
Thus, in the case of written input, the result of simplification is a text that features shorter sentences, simple grammar, and restricted vocabulary.
Honeyfield (1977) examined traditional simplification techniques for the preparation of language teaching materials, especially graded readers. He described the two principal forms of simplification: linguistic and content simplification.
Linguistic simplification in English involves “a process of de-transformation in which complex sentences are broken up into simple or compound sentences; nominalizations are resolved in separate sentences; tense relationships are standardized; modal meanings may be lexicalized; and anaphoric links are filled in” (Mountford, 1976, p. 151). Paraphrasing (e.g., replacing “wealthy” by “very rich”) is another technique used for linguistic simplification.
Content simplification involves rewriting a story in a simplified manner or omitting less important incidents or passages than the original version. According to Honeyfield (1977), “these processes produce material which differs significantly from normal English in the areas of information distribution — the way in which information is distributed in a text, — syntax and communicative structure — the way in which information is organized in a text for particular communicative purposes” (p. 431).
Simplifying syntax may lead to material lacking in cohesion, and hence, lead to resulting material that inadequately represents the semantic and rhetorical systems of normal English (Honeyfield, 1977). The following example extracted from Honeyfield (1977) illustrates a text that was simplified by avoiding the use of adverbial clauses and by using other rather simple adverbials (e.g., then, later, but):
We were rather worried about the ropes. We did not think about them during the day. We were too busy. But we thought about them during the night. We lay on mats in the cabin. Then we could both feel and hear the ropes. The logs moved under us. They were like an animal breathing. The first two nights were the worst. Later the water swelled the ropes. The ropes then held the nine logs together more tightly. But they still moved about. (Honeyfield, 1977, p. 435)
The passage lacks cohesion due to the fact that “the relationship of one piece of information to the next is often unclear” (Honeyfield, 1977, p. 435). Consequently, simplification of the language and content of written texts could induce learners to develop reading strategies that are inappropriate for unsimplified target language materials (Honeyfield, 1977).
Additionally, the use of limited vocabulary and short, simple sentences in simplified texts is likely to result in a broken up, unnatural discourse, which may differ significantly from authentic target language materials (Oh, 2001).
In a study conducted by Blau (1990) to investigate the effect of syntax, pauses, and speed on listening comprehension in learners of English as a foreign language and English as a second language, the author suggested that syntactic simplification (e.g., increasing the number of simple sentences) in the text did not increase language learners’ listening comprehension of texts read aloud. However, listeners’ comprehension was augmented by the inclusion of pauses at constituent boundaries in the aural text.
As mentioned by Yano, Long, and && Ross (1994), even though learners may comprehend a text from which all potentially unfamiliar linguistic items have been eliminated, this elimination prevents exposure to items that learners eventually should know.
Furthermore, Chaudron (1983) pointed out that simplified texts impeded rather than facilitated language learners’ comprehension of texts by creating an unnatural input that lacks natural materials needed to learn a language (e.g., implicitness, intertextuality, among other features of natural discourse).
As a consequence, as noted by Honeyfield (1977), language learners may not be able to fully comprehend the text, especially when asked to perform specific tasks such as inferring, which requires an understanding of those relationships (e.g., implicitness and intertextuality).
An alternative to the use of either authentic or simplified texts for language instruction is elaboration. As noted by Oh (2001), If one recognizes the need for a second/foreign language program to utilize some type of modified input to counteract learner deficiencies, efforts should aim to increase comprehensibility while maintaining essential features typical of unmodified input. In such efforts, elaborative modification represents a feasible alternative to simplification.
Elaboration can be defined as follows: Features such as slower speech, clearer articulation and emphatic stress, paraphrases, synonyms and restatements, rhetorical signaling devices, self-repetition, and suppliance of optional syntactic signals (e.g., relative and complement clause & markers) serve neither to simplify nor to ‘complexify’ the surface form, nor to create opportunities for interaction; rather, they are clarifications of meaning only, opportunities for the listener/reader to better decode the communication. (Parker & Chaudron, 1987, p. 110)
Elaboration, as described by Yano, Long, and Ross (1994), is a process in which unfamiliar linguistic items are offset with redundancy and explicitness. Elaborated materials are a result of several studies on “foreigner talk discourse” in the 1970s and 1980s, showing that in communicative interactions with nonnative speakers of English, native speakers of English often adjust or modify their speech in their attempt to make it more comprehensible to the second language listener (Long, 1983). Native speakers would slow the rate of delivery or use shorter utterances, but would not simplify the context of their speech in order to successfully communicate to non-native speakers.
According to Long (2007), speakers (native and non-native) would “negotiate for meaning” by using devices such as repetition, paraphrasing, confirmation checks, clarification requests, and several types of scaffolding represented by lexical switches, decomposition, etc.
Redundancy is another natural feature present in communicative interactions. Buck (2001) noted that language is redundant by nature, and there are so many clues to what the speaker is saying that listeners can understand even if speakers do not state it clearly. Speakers would instinctively modify their speech depending on the situation and their knowledge of the listener.
Buck (2001) observed that people would speak faster, run words together more, and be more indistinct when they share knowledge of a topic. However, speakers will speak more slowly and clearer when speaking to someone who has less background knowledge. The author concluded that comprehension takes place because language is so redundant that people do not need all the information to be clearly expressed in order to comprehend. We use our knowledge of the language to “replace” any missing information and construct meaning for ourselves.
As explained by Long (2007), elaborated texts (written or aural) can be designed by adding redundancy and regularity (also refer to as transparency) to a text, and often more explicit signaling of its thematic structure, followed by gradual removal of the modification provided as learner proficiency increases.
Redundancy, as noted by Long (2007), is achieved by repetition, paraphrase, provision of synonyms of low frequency lexical items, etc. Regularity, or transparency, is accomplished through parallelism, more frequent use of canonical word order, retention of optional constituents (e.g., subject pronouns in pro-drop languages), and matching order of mention to order of occurrence (e.g., The plane took off before the family reached the airport in preference to The family reached the airport after the plane had taken off, Long, 2007).
Thematic structure is defined as “any non-canonical word order that has the functional purpose of placing the known information first, and the new information second” (Parker & Chaudron, 1987, p. 115).
Thematic structure in English, as indicated by these authors, “is achieved syntactically by means of prepositional and adverbial phrase preposing, various types of cleft constructions and various types of extraposition” (Parker & Chaudron, 1987, p. 116).
The following example shows instances of signaling of thematic structure as presented by Parker and Chaudron (1987):
What separates the expert from the novice is the expert’s ability to remember board positions. This ability, it appears, is related to superior knowledge of the game, not to superior memory (p. 116).
Signaling of thematic structure was attained by the wh-cleft in the first sentence that separated the theme (which follows What) from the new information (which follows is).
The addition of an extraposition construction (it appears) reinforces the fact that the information that follows is new (Parker & Chaudron, 1987).
Elaboration of written texts
Parker and Chaudron (1987) studied the effects of elaborated input on language learners’ comprehension of academic discourse. They referred to elaborated input as a combination of features of redundancy and thematic structure. The authors found that although neither redundancy nor thematic structure showed a significant impact on improving learners’ reading comprehension, the higher reliability (Cronbach’s alpha) and greater correlation with other statistical measures of the elaborated passages indicated to be more natural, i.e., more coherent and better structured, than simplification.
This sort of results indicates that the elaborated reading text “was more like a normal reading passage possibly because it was more coherent and well-structured” (Parker & Chaudron, 1987. As noted by the authors, the high level of lexical and syntactic difficulty of the written passages could be responsible for the lack of effect of elaboration on learners’ reading comprehension.
Comprehension of the difficult syntax and complex vocabulary demanded too much processing and access to the target language grammar that learners could not take advantage of the modifications of thematic structures (Parker & Chaudron, 1987).
The authors concluded that elaborated input, which does not hinder comprehension but seems to be more natural, is a better choice for language instruction than non-elaborated texts. Yano, Long, and Ross (1994) investigated the relative effects of simplification and elaboration on 483 Japanese English learners’ reading comprehension. Participants read 13 written texts in one of the three versions: baseline, simplified, and elaborated, with length varying from a short paragraph to a two-page text.
Comprehension was assessed using 30 multiple-choice items. The results showed that language learners who read the linguistically simplified passage scored significantly better on a comprehension test than readers of the unmodified, original version of the same passage.
Readers of the elaborated version of the text also performed better than readers of the unmodified passages, but the difference in scores between the two groups was not statistically significant.
There was also no statistically significant difference between the scores of the students reading the simplified passages and of those who read the elaborated versions despite the fact that the elaborated texts were more complex in words per minutes and in words per sentence, about 50 % longer, and six grade levels harder in readability than the simplified texts.
A study conducted by Oh (2001) investigated the effects of two types of input modification, i.e., simplification and elaboration, on 430 Korean high school English learners’ reading comprehension. Reading passages in one of three forms (i.e., baseline, simplified, or elaborated) were presented to students who were divided into two proficiency levels: high proficiency and low proficiency.
She also examined the effects of modification type and learner proficiency on general, specific, and inferential comprehension processes. In Oh’s (2001) study, general comprehension items required the reader to grasp the main idea of a passage by combining seemingly unrelated pieces of information (e.g., finding the most appropriate title for a passage or judging the author’s attitude toward some passage content). Specific comprehension questions “required the reader to pay close attention to explicitly stated factual information in a passage in order to be able to identify the truth or falsity of specific propositions regarding the passage” (Oh, 2001, p. 78).
Inference items required the reader to draw implications from the text. The author hypothesized that “if elaboration is as effective as simplification for comprehension, it will constitute an alternative approach to written input modification because it allows more native-like target language input” (Oh, 2001, p. 91.
To construct the simplified version of the reading passage, low-frequency words were replaced by higher frequency words (e.g., credulous, coincidences, and obscure were replaced by believing, accidental events, and humble). In addition, multiword expressions were replaced by one word items with similar meanings (e.g., used to be, bring to a conclusion, and accept the fact were replaced by were, end, and believe), thus reducing the length of sentences as well.
In order to construct elaborated texts, Oh (2001) added redundancy and clearer signaling of thematic structure in the form of examples, paraphrases and repetition of original information, and synonyms and definitions of low-frequency words contained in the baseline passages.
Oh‘s (2001) major findings can be summarized as follows:
(a) simplified input, facilitated Korean high school English learners’ reading comprehension, although students of low proficiency did not significantly benefit from it;
(b) elaborated input significantly enhanced the overall reading comprehension of students at both high and low proficiency levels.
Learners who had read elaborated passages scored significantly higher on the comprehension test than did those at the same proficiency level who had read unmodified versions of the same passages. Surprisingly, those in the lower range of English proficiency seemed to do best on elaborated passages. For the high proficiency students, the facilitative effect of elaboration was comparable to that of simplification.
Oh’s (2001) findings also indicated no interaction effect between input modification type (e.g., simplified or elaborated) and learner proficiency. However, it is important to note that the high proficiency students benefited from input modifications to a greater extent than the low proficiency students did, meaning that modification of the passages had more of an impact on the reading comprehension of the higher proficiency group (Oh, 2001, p. 87).
Li, Xu, and Wang (2005) conducted a study, in which they investigated the effects of simplification and elaboration on 48 Filipino high school students’ second language reading ‘. comprehension. This study replicated Oh’s (2001) investigation by following her way of modifying and naming the passages: baseline version, simplified version, and elaborated version. Reading comprehension was evaluated based on students’ performance on general, specific, and inferential information. Three English reading passages in one of the three forms – baseline, simplified and elaborated – were presented to participants. Results indicated that elaborated written input was overall more comprehensible than baseline written input for Filipino learners of English; however, elaboration appeared more helpful for the low proficiency students than for the high proficiency students on general comprehension of the text.
Elaboration of aural texts.
Several studies have investigated the effect of elaborated language on listening comprehension. Chaudron (1983) conducted a study to investigate how different types of topic reinstatements affected second language learners’ recognition and recall of sentence topics in lectures. The topic reinstatements were repetition of the noun topic, rhetorical questions, synonyms, conditional clauses, and simple noun reiteration. The author wanted to examine which of the two devices would be more effective in promoting retention of the topic: syntactic simplicity or elaboration and redundancy.
Chaudron and Richards (1986) piloted a study intended to explore the effects of discourse signals and markers in “reading style” lectures (e.g., the lecturer reads from notes, or speaks as if he was reading from notes) on second language learners’ listening comprehension. The baseline version of the lecture consisted of a condensed written passage of a videotape on the expansion of the United States from thirteen colonies to an imperial nation presented to university English students.
The baseline version did not include any signals of discourse organization or linking between sentences other than what was necessary to convey meaning to the passage.
A second version of the lecture, the “micro” version, was constructed to include micro markers (e.g., markers of inter-sentential relations, framing of segments, and pause fillers). Inserted micro markers consisted of temporal links (e.g., then, and, now, after this), causal links (e.g., because, so), contrastive relationships (e.g., but, actually), relative emphasis (e.g., you see, unbelievably, of course), and framing/segmentation (e.g., well, OK, all right) among other links.
A third version, the “macro” version, contained signals of meta statements about the major prepositions within the lecture, or the important transition points in the lecture (e.g., what I am going to talk about today, let’s go back to the beginning, to begin with, etc.). The authors found that inclusion of macro markers signaling major propositions or the important transition points within the lecture enhanced listeners’ comprehension and retention of lecture information. However, micro markers signaling inter-sentential relations, ” framing of segments, and pause fillers did not aid the learners’ recall of the lecture information.
Chaudron and Richards (1986) noted that important information arises from this apparent differential effect of macro and micro markers. While the inclusion of micro markers are of less semantic value in the lecture information and they only allow the speaker extra time to plan the next utterance, the macro markers are explicit signals of the development of the lecture information. The authors explained that listeners learn to pay no attention to all the minor pause fillers and redundant inter-sentential connectors, and instead make use of the time to process the important parts of the text. The listener knows that paying attention to markers of the general organization of the text is a critical skill for the comprehension of the information expressed by the lecture. Chaudron and Richards (1986) concluded that their research has important implications for language instruction and material development. The authors pointed out that A lecture read from a written text will usually lack the kinds of macro-markers found in the more conversational style of teaching. A lecture that uses more macro-markers is likely to be easier to follow.