Summary:

Both L1 and L2 acquisition theory hold that children acquire language naturally by making hypotheses about language and testing them against the language in their environment. Formal language instruction is not necessary; language is acquired as children use it, and it emerges in a natural sequence. Caretakers provide comprehensible input as they adjust their speech to match the child's current comprehension. In order for language to develop, input must always be slightly above the present level of proficiency. At any given point, children do not acquire every structure available in the input. They note what is relevant to them at that stage of development and ignore the rest, showing a constantly developing grammar which gradually approaches that of the adult or native speaker. While L2 acquisition resembles that of L1, it differs as well. Grammatical structures are acquired in a different order and learners apply their L1 knowledge about language and the world to L2.

LANGUAGE AND LITERACY

First Language:

Educators have long recognized that oral language development has an impact on reading and that reading facilitates writing (Lipson & Wixson, 1991). From this relationship has stemmed the argument that LEP students should be introduced to listening, speaking, reading and writing in a sequential manner. But even in L1 acquisition, it is becoming increasingly clear that the impact is not one way. Teale and Sulzby (1989), proponents of oral language development in the classroom, point out the following:

Educators have long seen that a strong oral language base facilitates literacy learning. Furthermore, it is clear that children's developing reading abilities influence their writing. However, we also must recognize that reading experiences influence oral language... and writing actually improves children's reading skills.... For young children, the language arts mutually reinforce one another in development. (p. 4)

Children have not finished acquiring language by the time they enter school. Both syntax and lexicon continue to develop, aided by the acquisition of literacy. Tucker (1976), reviewing L1 development during the elementary years, concluded that "as the child begins to read, he may receive input from written language which contributes to increased sophistication in phonology, syntax, semantics and pragmatics" (p. 29). Nagy, Anderson & Herman (1987) estimate that children acquire as much as 3000 words a year through their reading. In one study, they found that "how easily a reader learns a new word depends in part on the degree to which the context immediately surrounding a new word gives information about that word; but far more important is the degree to which that concept requires the student to go beyond his or her current level of knowledge..." (p. 266). Translated into Krashen's (1982) terms, their findings indicate that vocabulary development requires that the reading provide comprehensible input and that it be at an i + 1 level. A similar study (Shu, 1995) found that out-of-school reading significantly influenced vocabulary acquisition for both American and Chinese school children, concluding that "incidental learning from context may be a universal in the written vocabulary development of children" (p. 88).

Syntax also continues to develop. In the early grades, the complexity of a child's oral language as well as his writing far outstrips the language he encounters in reading. Throughout the elementary years, however, syntax continues to develop (Gleason, 1989) and writing eventually becomes more complex than speech. Groff (1978) reports that by the middle school years, children's written language is more fluent and complex, with fewer errors, than their oral language. For speakers of non-standard dialects, written language is much closer to standard English. The disparity suggests that the increased complexity results from written rather than oral input. Edmaiston and Larsen (1983) did find that oral language proficiency differs significantly between high, medium and low third grade writers but caution against attributing causation, suggesting that writing and oral language mutually reinforce each other. The studies which led Pflaum (1974) to conclude that "instruction in vocabulary and oral language development during the preschool and elementary years will undoubtedly be important to reading growth" (p. 127) could also suggest that language proficiency is the result of rather than a prerequisite to reading. One study showed a positive relationship between sixth graders' oral language complexity and reading achievement. A similar study looked at children from kindergarten through 6th grade and found that from the end of third grade through sixth grade, subjects scoring high in language also scored high in reading. If oral language proficiency were a prerequisite, one would expect to find a strong relationship at the beginning. But if children acquire complex syntax and vocabulary through reading, one would expect the observed results. Stronger readers receive far more input and can be expected to acquire greater language skills than poor readers. This interpretation is supported by a recent study of precocious speakers (Crain-Thoreson & Dale, 1992) which questioned the dependence of literacy development on language skill, finding that verbal precocity at 20 months of age did not result in precocious readers but frequency of story-reading, i.e., exposure to print, at two years of age did predict both language ability and knowledge of print conventions over the next 2.5 years.

Ehri (1985) suggests that the impact of print on speech carries even into phonology. She notes that historically the establishment of a standard written form of words has resulted in shifts in pronunciation to match the orthography of the word and that the possession of a writing system seems to stabilize phonemic change. Based on her research, she proposes that children also alter their pronunciation to conform to the written form, especially speakers of non-standard dialects. She writes,

To summarize, it is suggested that written language supplies a visual-spatial model for speech and that when children learn to read and spell, this model and its symbols are internalized as a representational system in memory. The process of acquiring this system works various changes on spoken language, particularly at the phonemic and lexical levels....If children's pronunciations are nonstandard and if spellings symbolize standard pronunciations, print may teach children how to say these words. (p. 361)

That reading and writing influence overall language acquisition follows naturally from L1 acquisition theory. Children make hypotheses based on linguistic input provided at an understandable level. As the input and output are extended to written language, the process continues in that milieu (Gleason, 1989; Lindfors, 1987). While the process is not totally parallel - it is more deliberate and allows more time for thought -, nonetheless, reading and writing provide additional opportunities for children to experiment with language. Doehring and Aulls (1979) challenge the premise of readiness, claiming that the earlier a child begins to read, the more likely he is to acquire certain cognitive and language skills through reading or interactively with reading. They insist that "the idea that certain prerequisite skills must be developed to a high level before a child is `ready' to read seems theoretically untenable" (p. 30). Johnson (1985) applies the same arguments to disabled children for whom literacy acquisition is often delayed, suggesting that reading can supplement language remediation. Arguing that oral language difficulties may be the result rather than the cause of an inability to read, she summarizes,

Many language impaired students have difficulty in acquiring rules for, and facility in, oral language; such a difficulty can be improved through reading and writing activities. Because reading is a symbol system that can be presented simultaneously, rather than successively, it can provide a means of helping atypical learners perceive and abstract certain features they seem unable to learn from listening alone. Reading and writing also provide vehicles for the practice of rule application and for increasing automaticity (p. 67)

A number of studies stress the impact of literacy on the language acquisition of children with disabilities. Weeks (1979) cites several studies supporting early reading as a means of language development. One study involved profoundly deaf pre-school children with no formal language of any kind on which to base reading. The children were taught reading and sign language simultaneously. The children acquired language structures from their reading which were incorporated into both their signing and writing. By the end of the year, they showed a high motivation to read and write as well as functional language skills uncommon in severely hearing impaired children. Weeks conducted a similar study with a language-delayed preschooler. The child was already beginning to read, and Weeks and the mother started a reading program with her in hopes of helping her language development. Within a year, by age 4 years/ 9 months, the child's percentile rating in language was in the 90's.

Also challenging the concept of reading readiness is Williams (1992) who studied three profoundly deaf preschool children with diverse language but similar literacy experiences and found that while none had acquired a strong spoken language and only one a strong signed language, all had developmentally appropriate knowledge of literacy as a result of print exposure used to reinforce language acquisition. She notes that the children "often related meaning first to print and then to sign language and/or speech" (p. 11). She concludes,

The findings of this investigation suggest that there is no one pathway to becoming literate. While proficiency with verbal language is not a prerequisite to written language development, for many hearing children it is an avenue to literacy learning. In this investigation the opposite was true: Knowledge of written language became a pathway to spoken and/or signed language acquisition. (p. 10)

Another study (Rogow, 1993) examined the language development in a non-vocal child with severe multiple disabilities who, in spite of an inability to communicate orally or use her hands, was able to read and comprehend as well as write using special equipment in both English and Cantonese. An expert listener and avid reader, the child has strong receptive language skills and a command of language structure as demonstrated by a series of tests. Books have provided experiences and knowledge. Rogow stresses "the importance of literacy in the lives of children with multiple physical problems. Literacy not only provides access to rich language experiences, it also ensures the use of sophisticated communications technology" (p. 20).

The relationship between oral and written language is far more complex than can be explained by reading readiness theory. Children's oral language skills provide them with the vocabulary and understanding of grammatical structure needed for reading. Reading provides children with the complex syntax, vocabulary and style they will use in their writing. But writing also provides them with a knowledge of language which facilitates the reading process while interacting with written language develops the total language system. (Gleason, 1989; Lindfors, 1987; Ruddell & Haggard, 1985) The different language modalities interact with and build on each other and appear to be acquired simultaneously rather than sequentially. As Strickland and Morrow (1989) point out, "Oral language development begins before literacy and then parallels it. It supports literacy, but it need not be fully developed for reading and writing to begin" (p. 240).

Second Language:

In L2 acquisition, as in L1, reading and writing are increasingly being viewed as means for overall language development with L2 literacy developing in many ways like L2 oral language (Barnitz, 1985; Cummins, 1984; Krashen, 1982; McLaughlin, 1984). Reading as the receptive modality provides input from which the learner forms hypotheses about language. Krashen (1989, 1993, 1994), advocating that ESL teachers have students read for pleasure, notes that L2 learners, like native speakers, acquire vocabulary and spelling through reading. A study by Elley and Mangubhai (1983) supports his claim. Fijian children for whom English was the language of education but not of the community were randomly assigned to three treatments - Shared Book Experience, Sustained Silent Reading and the traditional means of English instruction, the Tate Oral English Syllabus. The first two treatments involved the use of high interest story books. While the two book experiences showed little difference in the long run, both resulted in considerably more gains in English than the traditional oral method. Elley (1991) found comparable effects in a series of studies on other "book floods" in elementary schools in the South Pacific. Elley notes:

In contrast to students learning by means of structured, audiolingual programs, those children who are exposed to an extensive range of high-interest illustrated story books, and encouraged to read and share them, are consistently found to learn the target language more quickly. When immersed in meaningful text, without tight controls over syntax and vocabulary, children appear to learn the language incidentally.... (p. 375)

Allen (1989) reports similar results from LEP children involved in a literature program. Of one child, she observes, "Not only did the sharing of folktales provide a form that supported this child as he wrote, it also gave him an input of literary language upon which he could draw. This influence can be seen in both sentence structure and vocabulary" (p. 61). Janopoulos (1986) also observed that L2 reading was related to greater language facility with his study which found that LEP students' pleasure reading in English correlated significantly with their English writing proficiency while being heavy readers in the native language showed no relationship with English writing.

L2 literacy acquisition may provide learners with more than vocabulary and syntax. Lado (1977) makes a point similar to that of Ehri's regarding L1 readers when he claims that learning the written form simultaneously with the oral will help students acquire the pronunciation while students often have difficulty distinguishing sounds with only auditory reception. Another consequence of literacy acquisition in both L1 and L2 is that children gain metalinguistic skills which allow them to analyze sentences and break phrases into individual words (Gleason, 1989). This process, known as segmentation, is important in learning a language's structure and results largely from exposure to print. Reading allows children to identify words as isolated units.

In using reading as input, the same factors apply as with oral language: the input must be comprehensible. Krashen (1989) states that "the Input Hypothesis (IH) assumes that we acquire language by understanding messages.... If IH is correct, it predicts, first, that more comprehensible input, aural and written, results in more language acquisition" (p. 440, 441). Citing a number of studies which found that L2 reading resulted in vocabulary development, he asserts that the emphasis, as with oral language, must be on meaning and that language is learned incidentally through use. The input, again like oral language, must be comprehensible, interesting/ relevant, not grammatically sequenced and in sufficient quantity (1982). Rigg (1986) reviewed the use of several passages purported to be equivalent in grade level with LEP children. The content, however, was different and students displayed considerable difference in comprehension according to their familiarity with the subject matter of the passages. She cautions that a readability formula cannot determine whether a passage is at a student's level; the material must make sense to the student.

Not only can LEP children read; they can write and through writing experiment with the language they are receiving. Hudelson (1986) lists several findings about LEP children and writing based on classroom research, among them: L2 learners can compose in the second language before mastering it orally; they can create different kinds of texts for different purposes; they can take readers' responses into account and make changes in their writing. As with both L1 and L2 oral language, children do not acquire all grammatical structures at once. They note from their receptive language those features most salient to their current level and try new hypotheses. Cummins (1984) stresses that "if children are exposed to a wide variety of written language and if they are allowed to continue to express themselves in writing, then errors will gradually approximate adult usage without explicit correction" (p. 243). Research supports this assertion. Seda and Abramson (1990) examined English writing emerging in both native-speaking and LEP kindergartners and found the same developmental process occurring in both sets of children. Dulay and Burt (1980) review several studies which report finding the same acquisition order in L2 writing as in L2 speech. One was conducted by Krashen and his students with 70 ESL university students of four language backgrounds and found a similar order to that found in oral studies of LEP adults. The other studies also dealt with adults and had equivalent results. Research on children also indicates that active hypothesis-making occurs in writing and that errors gradually approach those of the standard English speaker (Boyle & Peregoy, 1990; Lindfors, 1987). In order for this to occur, however, the child must be engaged in meaningful use of language. Lindfors writes of both oral and written L2 acquisition that "second-language learners who do rehearsal practice drills get better at doing rehearsal practice drills; but it is the second-language learners who engage in communicating in the second language who get better at doing that: communicating in the second language.... The second-language learner, like the first- language learner, is a child creatively constructing language in a social world" (p. 470).

If a second language is indeed acquired sequentially through listening, speaking, reading and writing, one would expect to see grammatical forms mastered orally before they appear in written language. If, however, written and oral language are acquired simultaneously with both listening and reading providing comprehensible input and both writing and speaking allowing the learner to experiment with and hypothesize about language, then one should find the learner becoming aware of the forms in listening and reading first and then experimenting with and gradually acquiring the grammar in both speech and writing. While it would prove difficult to determine which forms are becoming salient to the learner in the listening modality, the other three modalities can be subjected to error analysis in an effort to demonstrate, as is the purpose of this study, which of the two viewpoints is most convincing.

Summary:

Research has shown a strong relationship between oral and written language. This has been interpreted to mean that oral language is a prerequisite to literacy development. For L1 learners, the theory of reading readiness has been promulgated. For the L2 student, the delay of literacy instruction has been urged until a high level of oral language can be attained. For both, the assumption has been that language is acquired in a linear fashion, first listening, then speaking, reading, and writing. Subsequent research, however, has suggested that aspects of language are interrelated with each reinforcing the other. Much of the more complex L1 syntax and vocabulary is acquired after literacy instruction begins. For the L2 learner, reading provides additional input and writing provides further opportunities for hypothesis-making. Literacy and oral language can be acquired side-by- side. This is especially crucial for children who have no access to first language instruction and whose academic growth may be hindered by a postponement of literacy acquisition.

ERROR ANALYSIS

Error analysis is used in many fields in an effort to determine the cause of error and rectify it by analyzing the pattern of mistakes. For example, in a page of addition problems involving carrying, a random pattern of error might indicate that the learner could not add. A consistent shortage by ten, however, would indicate that the learner is forgetting to carry and tell the teacher that the concept needs to be retaught. Error analysis in the various areas of language development, though, is applied differently. Since language is seen as developing naturally with learners experimenting with intermediary forms of grammar before the adult or native speaker structures are acquired, error analysis concentrates on examining the nature of the development, and the changing constructs are not viewed as mistakes per se but rather as a series of steps toward the final grammatical form (Corder, 1981; Dulay, Burt & Krashen, 1982; Gleason, 1989; Gorbet, 1979; McLaughlin, 1984). According to Gorbet (1979), "the theory of error analysis proposes that in order to learn a language, a person creates a system of `rules' from the language data to which he is exposed; and this system enables him to use it....The basic task of error analysis is to describe how learning occurs by examining the learner's `output'" (p. 24).

The same analysis applied to written and oral output can also be applied to receptive language through reading miscue analysis. Goodman (1973) coined the term `miscue' for any deviation from the text. He proposes that, as with language production errors, the deviations are not random but indicate the reader's hypotheses about reading. Even the best readers demonstrate miscues as they read, most of which do not affect meaning or syntax (Wixson, 1979). Non-standard English speakers frequently substitute the form they use in their oral speech for the written form and such substitutions are not considered errors (Goodman, 1973; Leslie & Caldwell, 1990). The same grammatical shift to the reader's own grammar can be seen in LEP students, enabling the researcher to use miscue analysis to observe the development of a receptive modality. In the receptive modalities - listening and reading -, language learners do not take note of every feature but rather what has salience at their current stage of development (Corder, 1981). According to Goodman, Goodman and Flores (1979), "as they become bilingual, the readers will show this in their reading as they do in their speech. Their reading will reflect not only their first language but the extent to which they are coming to control English phonology, grammar, orthography, lexicon and idiom. If they are learning to read English while they are learning to speak and understand spoken English, their reading will both reflect and contribute to their growing control" (p. 31).

Error analysis assumes that children acquire language naturally by developing hypotheses about language. If the theory is true, then an error analysis of reading, writing and speech samples should provide a means of observing and describing the process and demonstrate whether L2 acquisition is occurring simultaneously across modalities.

INTEGRATED CURRICULUM

The reaction by educators against grammar drills and written exercises provided the impetus for the call to delay L2 literacy. Cognitive and academic development was being put on hold while students learned the second language. Teachers urged that learners be provided meaningful language experiences and opportunities to communicate in L2 rather than spend their time filling in workbooks. The philosophical pendulum swung to the opposite extreme, however, in urging that L2 instruction concentrate on oral development until the language was mastered. Those fortunate enough to have first language instruction could continue academic development while acquiring L2, but for the majority, literacy acquisition became delayed. This resulted from educators equating written language with the workbooks and exercises of traditional language instruction. The contribution which the written modalities could make to a rich language environment was overlooked. It was Jim Cummins (1984) who first articulated the difference between what he called BICS (Basic Interpersonal Communication Skills) and CALP (Cognitive/ Academic Language Proficiency). LEP children quickly acquire BICS and are perceived as proficient by teachers who then attribute subsequent academic difficulties to cognitive deficiencies. Cummin's research, however, shows that children entering school by age seven require an average of five to seven years to acquire grade equivalency in CALP, the abstract language needed to cope with academic tasks. This type of language is acquired largely through literacy. The literature shows a growing consensus that LEP children, especially those for whom L1 academic instruction has been unavailable, cannot afford to be delayed in their academic growth by a delay in literacy instruction (Cummins, 1984; Enright & McCloskey, 1988; Krashen, 1992; Mercado, 1991; Ovando & Collier, 1985; Short, 1989). Krashen (1992) writes that "when second language acquirers read for pleasure, they develop the competence they need to move from the beginning `ordinary conversational' level to a level where they can use the second language for more demanding purposes..." (p. 84).

The question becomes one of how children with limited English skills and perhaps little L1 literacy are to develop both oral and written language simultaneously. The key, according to proponents, is through integrated curriculum (Cummins, 1984; Enright & McCloskey, 1988; Ovando & Collier, 1985; Richard-Amato, 1988; Short, 1989). Integrated curriculum focuses all aspects of language on a task which must be meaningful and interesting to the learner. Typically, a unit topic (e.g., space) is chosen and all modalities are employed in studying the topic. Oral language activities flow from reading and writing which in turn are supported by oral interactions. Specific skill instruction is not incompatible, but the emphasis is on using the skills to accomplish a purpose, and skill instruction is accompanied by skill use. Enright & McCloskey (1988) list a number of theoretical assumptions on which the integrated model is based, among them:

1. The whole of language is greater than the sum of the parts. Language is a limitless capacity to make meaning and therefore should not be broken down and "taught" as tiny discrete "skills" or "habits" or "facts."
2. Students develop language and literacy through using language as a tool for creating and sharing meanings rather than through studying it as different "subjects"...
3. Comprehensible, interesting, and useful classroom language... is most helpful to students in their language and literacy development....
4. The whole of language development is greater than the sum of its parts. Students develop language and literacy as part of a broader process of semiotic or meaning-making development.... (p. 19, 20)

As reiterated by Krashen and others, the curriculum must be both meaningful and comprehensible. Integrating the curriculum allows for this. The oral language which develops as children are involved in meaningful activities provides a framework, a "scaffolding" (Boyle & Peregoy, 1990) which makes the reading more comprehensible and allows L2 readers to use the text cues which facilitate reading. A child who studies life in space and works with partners to design a space station develops the vocabulary and concepts necessary to read and write about the topic. When he is taught punctuation and then works with an editing group to make his space story publishable, skill instruction becomes meaningful. Cummins (1984) proposes that language proficiency be conceptualized along two continua, from cognitively undemanding to cognitively demanding and from context embedded to context reduced. Traditional instruction is both cognitively demanding and context reduced. While eventually LEP children need this level of proficiency, they cannot handle it until they develop CALP. The more cognitively undemanding the instruction, the more context reduced it can be. The more context embedded the instruction, the more cognitively demanding it can be. ESL instruction is often both context embedded and cognitively undemanding, delaying the acquisition of CALP. Integrated curriculum instructs simultaneously through all language modalities - listening, speaking, reading and writing - to provide the context which allows students to engage in cognitively demanding academic tasks while acquiring their second language.

SUMMARY

During the past few decades, as the theory of L2 instruction has developed, teachers have been recommended to delay the instruction of literacy skills on the grounds that children cannot benefit until they are orally proficient. Proponents of this delay note that children acquire L2 naturally as they do L1 and that children do not learn to read or write in their first language until they have acquired the language structures in their oral language. A body of research, however, indicates that literacy and oracy are interdependent even in the first language, with much of the language's complex structures and advanced vocabulary being acquired through exposure to print. For L2 learners, literacy developed along with the oral language allows each to reinforce the other and enables the development of CALP - cognitive/ academic language proficiency -, that more advanced and complex level of language needed for academic success. The current study has been proposed in support of this position with the hypothesis being that if children are indeed acquiring written and oral language modalities simultaneously, an error analysis should show language structures being acquired in a similar order across the modalities and should also show that structures were not mastered in the oral modalities before being present in the written modalities.

Go to Chapter 3