Multiple Embedded Scaffolds:
Support for English Speakers in a Two-way Spanish Immersion Kindergarten

Suzanne F. Peregoy
San Francisco State University

Owen F. Boyle
San Jose State University

Abstract

The present research focuses on opening activities in two bilingual Spanish immersion Kindergarten classrooms in northern California, Creekside and Seaside, in which similar numbers of native Spanish speakers and native English speakers were instructed primarily through Spanish. This study analyzes ways in which teachers structured opening activities to ensure comprehension and participation by children when the youngsters had little or no knowledge of the language of instruction, Spanish. Routines are analyzed in terms of scaffolding provided, with the concept of multiple embedded scaffolds proposed to explain how native English speakers are supported in language acquisition and content learning through a language that is new to them. Analysis revealed specific communication strategies such as functional repetition and the use of non-verbal cues to meaning that scaffolded comprehension. In addition, the teachers modeled language use themselves and structured events so that native Spanish speakers would model appropriate responses for non-Spanish proficient children to emulate. Finally, in the curriculum routines, teachers increased the conceptual and linguistic complexity of the content placed in the slots created by the routine, continually challenged children to higher levels of language performance, and engaged their participation in their zones of proximal development.

Sections of the Article

Introduction

Over the last 15 years, two-way immersion programs have proliferated due to their effectiveness in promoting particpants' language and literacy development in two languages while fostering positive cross-cultural attitudes (Lindholm, 1994). In the United States, where the target language is often Spanish, immersion programs are "two-way" when classrooms include balanced numbers of both native English speakers learning Spanish and native Spanish speakers learning English. The research discussed here focuses on two-way Spanish immersion Kindergarten classrooms. Drawing on findings from two observational studies, this article elaborates the idea of "multiple embedded scaffolds" (Peregoy, 1991) to describe how two teachers supported English speaking Kindergartners who knew little or no Spanish at the beginning of the school year.

The first study investigated how English-speaking Kindergartners faced the challenges of being instructed in Spanish, their new language, for all but twenty minutes of the four-hour Kindergarten day (Peregoy, 1991). Observations were made weekly from September to January and biweekly from January to June. Running narrative records were made on the spot and reviewed and elaborated at the end of the school day. For analysis, the narrative records were read and reread to identify patterns in the data, which were coded by categories that emerged. Results fell into two major categories: teacher-led events and child-initiated events. Findings were discussed in terms of environmental scaffolds erected by the teacher and children's responses as they quickly learned to participate in day-to-day classroom routines. In particular, the concept of multiple embedded scaffolds was proposed to describe the teacher's simultaneous use of several scaffolds to support second language learners' comprehension and participation.

The second study, replicating the design of the first, focused on second language learning opportunities and peer interactions in a two-way immersion Kindergarten that followed the same immersion model as the Kindergarten in the earlier study (Gundling, 1995). A similar participant-observation approach was used involving the collection of narrative records, which were analyzed qualitatively as in the previous study. Results were divided into teacher-led and child-centered events and reported in terms of (1) teacher scaffolding and (2) how and to what extent English speakers availed themselves of language learning opportunities by interacting with native Spanish speakers.

Using results from both studies, this article focuses on teacher-led events in the two Kindergarten classrooms, specifically Opening Activities, in order to further develop the idea of multiple embedded scaffolding (Peregoy, 1991). Teacher-led events were selected for further study because teachers in both studies were found to use a scaffolding structure that exhibited common elements, although the lesson content and implementation were quite different. We first review three theoretical bases underlying two-way immersion: comprehensible input, social interaction, and scaffolding.

Comprehensible Input

Language immersion programs, whether one-way or two-way, are based on the assumption that a second language is best acquired by being "immersed" in an environment in which the new language is used for natural communication purposes. Essential to success is the teacher's use of certain communication strategies that permit students to understand instruction well enough to participate and learn, even though they are still limited in second language proficiency. These strategies may include verbal accommodations such as paraphrasing and repetition, and the use of non-verbal cues to convey meaning, such as concrete objects, visuals, demonstrations, dramatizations, and modeling. When the use of these strategies is successful, the teacher's instructional language serves as "comprehensible input" (Krashen, 1982; Wong Fillmore, 1985), providing the raw material from which children can begin to construct a working knowledge of their new language.

Social Interaction

Essential though it may be, comprehensible input is not sufficient for language acquisition (Swain, 1985; Long & Porter, 1985). Learners also need to be able to respond, verbally or non-verbally, in ways that keep them in the conversation, where meaning is negotiated and communication is accomplished through give and take, trial and error. During conversational interactions, language learners receive important feedback from native speakers as to how effectively they are using their new language—feedback that will help them further develop and refine their budding language knowledge. Furthermore, language learners can ask their proficient conversational partners for repetitions, rephrasings, and word meanings, thus prompting their interlocutors to adjust their talk for the language learner's particular needs. In two-way immersion, the teacher is not the only proficient speaker with whom the children interact. Rather, the inclusion of native speakers adds natural opportunities for children to interact and communicate with native speaking peers, thereby multiplying and enriching natural language learning opportunities.

In classrooms, teachers structure social interactions in a variety of ways at the service of communication and learning: teacher to whole class, teacher to small group, teacher to individual child, children in pairs or small groups, and so forth. Each of these participation structures offers different kinds of language learning opportunities in terms of the content, the communicative purpose, the complexity of talk needed, and the extent to which meaning can be conveyed by context and non-verbal cues (Wong Fillmore, 1985). For example, the participation structure in which a native Spanish-speaking child builds blocks with a limited Spanish proficient child differs from the participation structure in which the teacher leads a story discussion with the whole class. Each generates its own language learning opportunities as meaning is negotiated, e.g., where to put the next block vs. the question of whether Goldilocks should have entered the bear's house when no one was home. The blockbuilding event is less complex in content and language than the discussion of Goldilocks, but offers important language learning opportunities specific to the child's own communicative purposes and individual level of interest and development. In two-way immersion, the learning environment is structured to optimize communicative success between teacher and children and among the children themselves through social interactions, ranging from formal teaching-learning interactions to informal peer interactions.

Scaffolding

Together with social interaction and comprehensible input, scaffolding forms the third theoretical basis for language immersion programs. Based on Vygotsky's (1978) seminal work, scaffolding may be defined generally as temporary support or assistance, provided by someone more capable, that permits a learner to perform a complex task or process that he or she would be unable to do alone. Scaffolded performance, repeated and supplemented with direct instruction as needed, eventually leads to full development (Cazden, 1983). The scaffold may then be dropped and a more advanced or difficult task taken on. Scaffolding is therefore dynamic, changing over time to accommodate and challenge learners' advancing zones of proximal development (Vygotsky, 1978). In two-way immersion, students' comprehension and participation must be scaffolded so that they can understand, interact, communicate, and learn despite their limited proficiency in the language of instruction.

The scaffolding metaphor has been discussed in a variety of ways across various disciplines. Here we are looking specifically at scaffolding that supports language acquisition, especially second language acquisition in Kindergarten. We begin with Bruner's seminal discussion of mothers' conversational scaffolds with their toddlers (Bruner, 1978). For example, if the child says: "Mommy birdie," the mother might reply, "Yes, there's a little bird on the fence! Where did he come from?" In this way, the mother takes the child's topic, elaborates it in a natural conversational manner, and asks a question that encourages the child to continue the conversation. In this way the mother has assisted the child in taking part in a conversation at a level the child would be unable to achieve without assistance. At the same time, she provides a model of mature linguistic usage that serves as input for the toddler's language acquisition.

Following Bruner's lead, the scaffolding metaphor has been applied to literacy development (Boyle & Peregoy, 1990). Literacy scaffolds are defined as instructional strategies that provide built-in teacher or peer assistance that permits learners to read or write texts at a higher level than they could achieve unassisted. When using dialogue journals, for example, teachers scaffold emergent literacy in much the same way that mothers scaffold toddler's conversations. Dialogue journals represent written conversations in which the teacher's reply to a child's journal entry elaborates on the child's topic while modeling mature spelling and usage. If the teacher ends with a question, the child is encouraged to write more, thereby keeping the conversational ball rolling. Additional examples of literacy scaffolds include predictable storybooks, pattern writing, mapping, and reader's theater (Peregoy & Boyle, 1997).

Another useful view of scaffolding is provided by Snow (1977), who points out that routines provide scaffolds or structures wherein gradually more complex content may be implemented to challenge the learner to his or her next level of development. Snow refers to this practice as "upping the ante." Although Snow's research dealt primarily with mothers' communication with their infants, applying her idea to school routines provides teachers and researchers with a deeper appreciation of the power of routines in schooling. In Snow's words: "We think of routines as simple and unsophisticated . . . but their simplicity allows for the introduction, into slots created by the routine, of fillers considerably more complex in structure and/or content, than could possibly be dealt with elsewhere" (in Cazden, 1983, p. 9). The routines themselves scaffold performance at a higher level.

Using these scaffolding concepts as analytical tools, the remainder of this article describes scaffolding during opening activities in two Spanish immersion Kindergartens.

The Kindergarten Classrooms at Creekside and Seaside

The two Kindergarten classrooms, to be referred to as Creekside and Seaside, were located in different semirural areas within 60 miles of San Francisco, CA. Adhering to the same two-way immersion model, both classrooms served balanced numbers of native Spanish speakers and native English speakers, with a total of 30 children in Creekside (47% native Spanish speakers and 53% native English speakers) and 31 children in Seaside (39% native Spanish speakers and 61% native English speakers). During the 8:30 a.m. to 12:00 p.m. Kindergarten day, instruction was delivered in Spanish for all but the 20-25 minutes devoted to English language arts. Both classrooms followed a similar scheduling plan, with the first 30-45 minutes devoted to opening activities led by the teacher with the whole class seated on the rug. Opening activities included calendar, songs, finger plays, and story read alouds. At the end of the opening activities, the teacher would explain the activities prepared at each of several centers, variously set up with math manipulatives, trade books and books on tape, blocks, dramatic play items, and art materials. After opening activities, the remainder of the day consisted of children rotating through each center in small groups. After center time, children went to another teacher for English language arts and reassembled again for closing time in Spanish. Two kinds of participation structures predominated during the Kindergarten day: a teacher-directed participation structure in which the teacher led whole class or small group activities, and a child-centered participation structure in which small groups of children were left on their own to carry out tasks set up by the teacher at tables and centers.

Previous analysis of both teacher-directed and child-centered participation structures (Philips, 1983) revealed different scaffolding and language learning opportunities inherent in each. Here we focus on the teacher-directed opening activities. In analyzing the opening activities in both classrooms, we describe how both teachers carried out clerical and managerial tasks in similar ways, while implementing two different curriculum routines: the storybook routine at Creekside and the Oso routine at Seaside.

Getting Organized During Opening Activities: Creekside and Seaside

Observations revealed that opening activities in the two Kindergarten classrooms served three main classroom functions: clerical (taking roll and getting lunch count); managerial (reviewing class rules and explaining center activities) and curricular (working with letters, numbers, stories, and other curriculum content). The following observation at Creekside, for example, reveals how these functions were smoothly carried out at the beginning of the school year, when nearly half of the students were non-Spanish speaking:

Opening activities at the rug required children to adhere to certain behavioral rules such as hand-raising for a speaking turn and sitting still to maximize listening and minimize distraction. These rules were taught entirely in Spanish, even at the beginning of the year when many children spoke no Spanish. At the beginning of the school year, for example, the teacher would review classroom rules each day entirely in Spanish. For each of her classroom rules, she had posted a small symbolic picture on the bulletin board. For example, for the "talk quietly" rule, there was a picture of an index finger over a pair of lips, with "shhh." For the "raise your hand" rule, she had posted a picture of a hand waving in the air. Not only did she review the rules, she also enforced the rules using only Spanish. Early in the year, for example, a little English-speaking boy was sent to sit at the table for behaving disruptively during opening activities. The teacher said to him, "Yo creo que vas a sentar mejor en la mesa. Vete a la mesa." ("I think you are going to sit better at the table. Go to the table.") The boy did so, comprehending the general idea more than the specific words spoken. Daily routines such as these offered natural opportunities for certain words and phrases to be repeated in context and thereby learned quickly by all children.

The language that both teachers used during clerical and managerial duties was supported by various non-linguistic cues to meaning such as pictures, gestures, and dramatizations. As a result, children were able to understand the routines and respond appropriately. At the beginning of the year, for example, the teachers structured questions so that second language learners could answer non-verbally or else select from two choices provided by the teacher, e.g., "¿Quieres el rojo o el verde?" ("Do you want red or green?") In addition, the native Spanish speakers, who naturally caught on to the routines more quickly, served as performance models for the English speakers. Both teachers built in ways for the Spanish speakers to model for the English speakers. Thus, within the overarching scaffold, the routine itself, teachers used specific communication strategies such as repetition of key phrases accompanied by picture and gesture cues to help Spanish learners understand and participate in the opening activities. In addition, teachers structured events so that native Spanish speakers could model the appropriate verbal responses. These communication strategies represent additional scaffolds embedded in the overarching scaffold, the routine itself. While the clerical and managerial routines at Creekside and Seaside were similar, the curriculum routines during opening activities in the two Kindergartens were very different. The Creekside teacher employed a storybook routine, while the Seaside teacher used the class mascot, a stuffed bear, to engage students in a routine dealing with health and hygiene. Each of these routines is now discussed.

The Storybook Routine at Creekside

The Creekside Kindergarten teacher typically presented a storybook during opening activities as part of her language arts curriculum. At the beginning of the year, for example, she used books in Eric Hill's "Spot" series, which were available in Spanish. These are simple, predictable picture books with one sentence per illustrated page telling about a dog named Spot. By the end of the year, the teacher had introduced more complex stories, such as "The Three Bears" and other familiar fairy tales involving more complex plots than the Spot stories. Her daily read aloud provided various opportunities to engage all children in understanding and enjoying the story, with pictures supporting children's comprehension. The teacher would reread the same story several days in a row, varying the presentation, and encouraging children to chime in as they began to memorize the story. The story then formed the basis of activities for children to engage in at center time. For example, computer generated copies of the story were available for each child to illustrate at the bookmaking center, where about five children would work at a time. Children were often observed "reading" their books to each other, usually based partly on memorization and partly on children's sight word recognition and sound/symbol knowledge. As they read to each other, children practiced their emergent reading skills in their new language, Spanish. Interestingly, the print itself often jogged their memory and supported their oral rendering of the story. In this way, written language scaffolded oral performance in their second language. Finally, the children could take their own computer copies of the story home to read to their parents.

The Oso Routine at Seaside

The Seaside teacher used a curriculum routine with the class stuffed bear, Oso, to develop concepts about health and hygiene in ways that gradually increased children's verbal and non-verbal participation (Gundling, 1995). On the first day of school, the teacher greeted the children at the door while holding Oso, who patted the children on the head and said, "Buenos Días." During the first week of school, the teacher invited children to participate in a movement activity in which the teacher called out, "Arriba, abajo, al lado, al lado." ("Up, down, sideways, sideways.") While calling the commands, the teacher moved the bear so it modeled the response, a physical response reminiscent of the Total Physical Response method of language teaching (Asher, 1982). The teacher then made Oso do sit-ups, while counting to 10. Children watched and counted with the teacher if they wished. In the first weeks, non-Spanish-speaking children were able to participate in a meaningful activity through non-verbal responses modeled first by Oso and later by the Spanish-speaking children.

By the end of September, the Seaside teacher had introduced five health and hygiene concepts into the Oso routine: washing, exercising, eating well, getting up early, and brushing teeth. Each concept was represented by a picture on a board labeled in block print, "¿Que debemos hacer para mantenernos con buena salud?" ("What should we do to stay healthy?") The teacher set up questions to ask Oso in Spanish for each item, e.g., "¿Te bañastes, Osito?" ("Did you bathe, Osito?") At the beginning, the teacher asked Oso the question, and a child was selected to answer sí or no for the bear. Soon, children volunteered to ask the routine health question while a second volunteer would answer sí or no. The comprehension of the health question was supported by the picture, while children's verbalization was supported through teacher modeling. In addition, if an English-speaking child stated the question in English, the teacher would repeat the question in Spanish, thereby modeling the preferred response. As the year progressed, the teacher would move on to the next health question. As she did so, she increased the complexity of language required for verbal participation. Thus, while the first health question required only a yes or no response, later questions required more elaborate linguistic responses. By March, for example, if Oso had not slept well, the teacher would ask the children why. Children might reply that he had stayed up too late the night before, or that he had nightmares, responses that required more complex language.

As the year progressed, the teacher added increasingly complex verbal and dramatic elements to the routine for each question. For example, when asking if Oso had eaten breakfast, she would instruct a volunteer to bring some food from the dramatic play center to feed Oso. First the teacher would name each food item. Then she would ask the bear whether he liked each one using the simple sentence pattern, "¿Te gusta . . . ?" ("Do you like . . . ?"). In later weeks after modeling the routine several times, the teacher would select a volunteer to ask the question, with another volunteer answering si or no. The teacher was quick to introduce humor into the routine as well. For example, if Oso said he had not bathed, the teacher would hold her nose and make a face of disgust, while exclaiming, "¡Qué cochino!" ("How filthy!"), much to the delight of all the Kindergartners.

Figure 1 illustrates the multiple embedded scaffolding components inherent in the Storybook Routine and the Oso Routine. The scaffolding components include the curriculum routine, comprehension supports, teacher modeling, peer modeling, the sequence in which language learners gradually assume responsibility for the response, and the manner in which the teacher increases the complexity of language and content of the routine. Figure 1 shows how two different curriculum routines followed a common structure composed of multiple embedded scaffolds that permitted language learners to understand and participate in lessons even when their proficiency in the language of instruction was limited.

Conclusion

We have described the multiple scaffolds that teachers embedded into opening activities serving clerical/managerial and curricular functions. At the beginning of the year, routines such as roll call and classroom rule enforcement were scaffolded to help second language learners participate appropriately. These simple routines provided initial language learning opportunities while integrating language learners into the Spanish language environment. The content of these routines remained stable throughout the year. In other words, the teachers did not "up the ante" in these routines. In contrast, in the two curriculum routines, story reading and Oso, the teachers provided content and accompanying language that increased in length and complexity as the school year progressed. The Creekside teacher made use of predictable storybooks that increased in length and complexity as the year progressed, while the Seaside teacher made creative and ingenious use of the Oso routine to introduce health and hygiene concepts that became more elaborate in content and language as the year progressed.

Data analysis focused initially on single instructional events to determine how teachers scaffolded non-Spanish proficient children's engagement and performance at a given point in time. Although one can identify multiple embedded scaffolding at any given moment, its dynamic evolution over time is what bears true fruit in language acquisition and content learning, thus the need to examine scaffolding over time. The usefulness of routines as scaffolds comes into play as a result of their repetition over time. Time, verbal explanation, and mental energy are saved when routines are in place, and those savingscan be invested in helping children grapple with the increasingly complex content offered during those routines. Over time, the teachers increased lesson complexity in language and content, by upping the ante and challenging children to their next level of development. It is also over time that the teachers gradually relinquished their lead role to the children as they became capable of taking over the task, whether reading a predictable story or answering elaborately for Oso.

FIGURE 1: Multiple Embedded Scaffolds for Storybook and Oso Routines

Routine/Scaffolding	Storybook Routine	Oso Routine
Daily routine	A predictable storybook is read aloud each day.	The class bear, Oso, is involved in a health and hygiene lesson each day.
Comprehension supports	1. Storybook contains picture cues, repetitive refrains, predictable episodes. 2. Teacher calls children's attention to meaning of cues while reading story.	Teacher presents health and hygiene concepts using pictures, gestures, dramatic play objects, and Oso's physical actions.
Teacher models	Teacher models language and reading behaviors while reading book.	Teacher models questions and commands; teacher speaks for Oso modeling responses to questions/commands.
Proficient peers model	1. Children chime in on refrains andother story parts. 2. Children read story to each other at centers.	Children begin to play teacher's part by asking questions/giving commands; children answer questions for Oso.
Sequence in which responsibility for response is gradually assumed by language learner	1. Teacher reads/children listen and look. 2. Teacher reads/children chime in 3. Children read story to each other 4. Children read story to parents (independent)	1. Teacher operates the stuffed bear, asks questions, answers questions for Oso. 2. Children do actions; begin to answer for Oso. 3. Children take over teacher's role of asking questions; other children answer.
Complexity of language and content increases over time	Teacher uses books of gradually increasing difficulty in length, vocabulary, content, predictability, and picture support.	Children's responses go from non-verbal, to one word, to several words, as questions require.

Implications for Research and Teaching

This study leads to several implications for research and teaching. First, additional research is needed on teacher scaffolding during teacher-led instructional activities to determine the extent to which multiple embedded scaffolding is used, not only in Kindergarten, but in other grades as well. Further research is also needed to ascertain the efficacy of multiple embedded scaffolding for language acquisition, content learning, and social participation among second language learners of varying L2 proficiency levels. Teachers themselves might evaluate the model by implementing it in their classrooms and evaluating the outcomes in terms of student participation and learning. This type of teacher research could lead to greater elaboration and refinement of the concept of multiple embedded scaffolding. Validated and refined, the model would be useful for teachers of young second language learners to use as the basis of instructional planning over the course of the school year.

References

Asher, J. (1982). Learning another language through actions: The complete teacher's guidebook. Los Gatos, CA: Sky Oaks Production.

Boyle, O. F., & Peregoy, S. F. (1990). Literacy scaffolds: Strategies for first-and second-language readers and writers. The Reading Teacher, 44 (3), 194-200.

Bruner, J. (1978, September). Learning the mother tongue. Human Nature, 43-48.

Cazden, C. (1983). Adult assistance to language development: Scaffolds, models and direct instruction. In R. Parker, & F. Davis (Eds.), Developing literacy: Young children's use of language (pp. 3-18). Newark, DE: International Reading Association.

Gundling, L. (1995). Second language acquisition and sociolinguistic interaction in a two-way Spanish immersion kindergarten. Unpublished Master's Degree Field Study. San Francisco State University, San Francisco, CA.

Krashen, S. (1982). Principles and practices in second language acquisition. Oxford: Pergamon Press.

Lindholm, K. (1994). Promoting positive cross-cultural attitudes and perceived competence in culturally and linguistically diverse classrooms. In R. A. DeVillar, C. J. Faltis, & J. P. Cummins (Eds.), Cultural diversity in schools: From rhetoric to practice (pp. 189-206). New York: SUNY Press.

Long, M., & Porter, P. (1985). Group work, interlanguage talk, and second language acquisition. TESOL Quarterly, 18, 207-227.

Peregoy, S. (1991). Environmental scaffolds and learner responses in a two-way Spanish immersion kindergarten. Canadian Modern Language Review, 47 (3), 463-476.

Peregoy, S., & Boyle, O. (1997). Reading, writing, & learning in ESL: A resource book for K-12 teachers, 2nd ed. New York: Addison Wesley Longman, Inc.

Philips, S. U. (1983). The invisible culture: Communication in classroom and community on the Warm Springs Indian Reservation. White Plains, NY: Longman.

Snow, C. (1977). The development of conversation between mothers and babies. Journal of Child Language, 4, 476-56.

Swain, M. (1985). Communicative competence: Some roles of comprehensible input and comprehensible output in its development. In S. M. Gass, & C. G. Madden (Eds.), Input in second language acquisition (pp. 235-253). Rowley, MA: Newbury House.

Vygotsky, L. (1978). Thought and language. Cambridge, MA: MIT Press.

Wong Fillmore, L. (1985). When does teacher talk work as input? In S. Gass & C. Madden (Eds.), Input in second language acquisition (pp. 17-50). Rowley, MA: Newbury House.