-3-

Linguistics

Theoretical Linguistics

Applied Linguistics

Sentence Complexity and Clause Subordination in Children's Spanish

Despite the growing interest of the past two decades in language acquisition among Spanish speakers, the development of clause subordination within complex sentences in the Spanish of children has received very little attention to date. Often development of subordinate clauses in Spanish has been considered only peripherally, i. e., as it relates to children's use of indicative or subjunctive mood within complex sentences. Studies which have treated children's use of subordination in Spanish consist primarily of doctoral dissertations, many of which have attempted comprehensive descriptions of the syntactic development of Spanish-speaking children for a particular geographic area. While such exploratory studies have contributed to the developmental literature for Spanish, most have not directed their focus to specific areas of syntactic or semantic development, such as the topic under review. Use of subordination and complex sentences among Spanish speaking children has rarely been explored as an area of syntactic or semantic development in its own right. To my knowledge, in-depth and systematic investigation of this area of language development among Spanish-speaking children has not appeared in the published literature to date.

The purpose of the present paper¹⁴⁶ is to review the findings reported in studies of syntactic development in Spanish as they relate to children's use of subordinate clauses within complex sentences. Studies treating bilingual Spanish-speaking children will be reviewed first, then those dealing with monolingual children. Ultimately, an attempt will be made to summarize such findings as have emerged in the developmental literature to date and to suggest directions for future research into this particular and most significant area of language development among Spanish-speaking children.

One of the major studies to date of the acquisition of grammatical structures by

––––––––   489   ––––––––

Spanish-speaking children is that by Gustavo González (1971), who attempted to establish tentative norms regarding developmental stages of Spanish among Mexican-American children in Texas. González studied the linguistic performance of 27 children from the Brownsville area, with three subjects representing each of nine age intervals from 2.0 to 5.0 years old. González observed development of sentence complexity and children's use of noun, adjective, and adverb clauses.

María Brisk (1972) attempted to determine the level of syntactic development of bilingual New Mexican children, interviewing seven five-year olds, two from urban Albuquerque and five from a nearby rural area. Among the structures Brisk observed were children's complex and compound sentences. Although the relevant discussion is brief, the observations are of interest.

Maryann McKay (1975) studied the effect of grade on the production of Spanish syntactic structures among 96 Mexican-American children in grades one through four in the San Francisco Bay area. McKay considered syntactic structures including Complex T-units (i. e., one main clause plus one or more subordinate clauses), Noun, Adjective, and Adverbial Clauses, and Total Number of Dependent Clauses.

Barbara Merino (1976), in another study of the San Francisco area, investigated developmental trends in 41 bilingual Chicano children aged five to eleven in grades kindergarten through four. Merino considered the effect of grade on both children's comprehension and their production of several syntactic structures, three of which, (i. e., «Subjunctive», «Conditionals», and «Relatives»), involved complex sentences.

Among the few studies which have treated the syntactic development of monolingual Spanish speakers is that by Samuel Gili Gaya (1972) who, in a study undertaken prior to 1960, investigated linguistic development of preschool, first-grade and fourth-grade children in Puerto Rico. Among his observations are many which relate to children's use of subordinating conjunctions and various types of modification expressed in subordinate clauses.

In a study of late stages of syntactic development among 55 monolingual Chilean children ages six to ten, Max Echeverría (1975) investigated the comprehension of several structures, including 1) subjunctive or conditional after decir [to tell], and 2) relative clauses.

Robert Blake (1980) investigated mood selection among 135 monolingual Mexico City children ages four to twelve. The types of contexts considered included adjective and adverbial clauses as well as various semantic categories involving noun clauses.

Explicit attention to sentence complexity and subordination in the developmental literature regarding Spanish-speaking children has been very limited. Before considering the studies as they relate to subordination of specific types of clauses, we note first some general observations made by the investigators regarding subordination within complex sentences in Spanish. Brisk (1972), observing complex sentences among New Mexican children, noted that subordination, in general, was «still very much in the process of development» among her five-year old informants (116). Similarly, Merino (1976) noted that among San Francisco children, the categories of «Subjunctives», «Conditionals» and «Relatives», all of which involved complex sentences, were among the most difficult structures, with significant differences observed by grade for children's production of subjunctives and conditionals (168). On the other hand, McKay (1975), in her study of San Francisco children, found no significant effect of grade on children's production of Complex T-units, noun, adjective, and adverb clauses nor on total number of dependent clauses. In attempting to explain this pattern of no growth, McKay (94) suggested that the Spanish-dominant bilingual children had already acquired Spanish before entering school. Gili Gaya, in his study of Puerto Rican children, noted their use of various subordinating conjunctions. Such forms, he suggested, have more a logical than a lexical function, and that grammatical function is defined gradually in the mind of children with continued use and practice (138).

Use of complex sentences by Spanish speaking children has often been discussed only or primarily as it relates to children's use of subjunctive mood. Gili Gaya (1972) observed use of subjunctive among preschool, first-grade and fourth-grade children in

––––––––   490   ––––––––

Puerto Rico. Fortunately, he also observed their development of various types of semantic modification, especially adverbial, and the use of subordinating conjunctions in that regard. Gili Gaya's observations will be noted in the following pages as relevant. Blake (1980), in a study of Mexico City children's selection of indicative or subjunctive mood, considered syntactic contexts of adjective and adverbial clauses and various semantic categories involving noun clauses. Although Blake found developmental patterns regarding the use of subjunctive within such contexts, unfortunately neither the development of subordination as such nor the development of specific subordinate clause types in the Spanish of the monolingual Mexican children was treated explicitly. Merino (1976), in a study of bilingual San Francisco children, included three different contexts involving subjunctive mood in complex sentences; unfortunately, however, her results were reported only «for subjunctive» as a category, rather than for the specific types of clauses included therein.

To the extent possible, the findings in the developmental literature will be considered below as they relate to specific types of subordinate clauses classified according to their grammatical function, i. e., nominal, adjectival, or adverbial, within the complex sentence. Those studies treating bilingual children will be reported first, then those dealing with monolingual Spanish-speaking children.

Noun Clauses. González (1972) reported that the use of noun clauses by Texas children varied according to the specific grammatical function the noun clause served within the matrix sentence. Noun clauses functioning as direct object after decir [to tell] (e. g., «Dime dónde está» [Tell me where he is]) were introduced by children aged 2.6 (31), were used frequently by the 3.3 year olds and older (76), and were used with verbs such as querer [to want] and saber [to know] with indicative or subjunctive mood correctly distinguished after age 4.0 (109). Relator words introducing object noun clauses at age 3.6 were que [that], dónde [where], si [if], and cómo [how] (90); at age 4.0 quién [who] was included among the relators to introduce such clauses (109). González noted that noun clause used as predicate after ser (e. g., «Este es el que lo checa» [This is the one who checks it]) (76), although introduced at age 3.3, was not observed again until age 4.6, and then only sporadically. Regarding noun clause as sentential subject, González reported only isolated instances at age 3.6 (90) and 4.0 (109); no use, however, was reported for children at later age levels. González observed a similar construction which he considered separately and described as the «es que [it's that] + sentence» construction or pattern, which would seem to involve a noun clause as subject. In any case, he observed an isolated case of this structure at age 3.3 (77) but more frequent use at age 4.0 and 5.0 (108, 137).

Brisk (1972), in her study of New Mexican five-year-olds, observed that among the rural children, noun clauses functioning as direct object were introduced by a «variety of relaters», (i. e., que, donde, como, and cuando [when]) after verbs such as querer, decir, and saber (78). She noted the less frequent use of this construction by the urban informants. When Brisk elicited among the urban children for repetition of a sentence involving noun clause after querer in a volitive context with change of subject (i. e., «Mi mamá quiere que yo vaya» [My mother wants that I go/My mother wants me to go]), she noted that they changed the subjunctive verb in the dependent clause to an infinitive (i. e., «Mi mamá quiere ir» [My mother wants to go]). Noting the urban children's infrequent use of noun clauses even when elicited, Brisk concluded that «the process of noun-clause subordination has not been fully acquired by these children» (79). She observed that noun clauses functioning as subject of «impersonal verbs» such as parecer [to seem] and ser [to be] (e. g., «Es que el negro juega mucho» [It's that the Black boy plays a lot]) were used rarely, and only by the rural New Mexican children (79, 116).

McKay (1975) found no significant effect for grade on San Francisco children's production of object noun clauses; frequency was very high among informants at all grade levels (82).

Merino (1976), in a study of the same Bay area, included in her category of three «Subjunctive» items one sentence involving object noun clause after querer in an «optative» environment (i. e., «La niña quiere que lave la ropa» [The little girl wants X to wash the clothes]) (271). Merino found that children's comprehension showed a developmental trend over grade, with older children generally out performing the younger (173-74); the differences, however, were not statistically significant.

––––––––   491   ––––––––

She reported that Spanish «Subjunctive» was one of the «most difficult» categories for subjects of all grades to understand (155). Differences in children's production of «Subjunctive» across grades were statistically significant (132). Unfortunately, however, Merino reported her findings only for «Subjunctive» as a category; no results were reported for children's comprehension or production of the specific noun-clause context noted above.

Among the studies of monolingual children's Spanish, very little mention has been made of the use of noun clauses. Gili Gaya (1972) cited examples of object noun clauses introduced by verbs of volition (e. g., querer and decir) among preschool Puerto Rican children. The subordinating conjunction que was considered to be well established already among the four-year old children (e. g., «Le dice que juegue» [He tells him to play] (60, 65).

Echeverría (1975), in his study of Chilean children, investigated the comprehension of two types of object noun clauses: 1) decir introducing subjunctive in a volitive context (e. g., «Charlie le dijo a Lucy que le comprara un helado» [Charlie told Lucy to buy him an ice cream]), and 2) decir introducing conditional in contexts of reporting (e. g., «Charlie le dijo a Lucy que le compraría un helado» [Charlie told Lucy that he would buy her an ice cream]) (116). Comprehension of structures such as 1) can be explained, as Echeverría noted, by the «Minimal Distance Principle»¹⁴⁷, which states that the noun phrase which most closely precedes the complement verb is assigned as subject of that verb (73). In example 1) above, «Lucy» would correctly be interpreted as subject of the complement verb comprara. Sentences with decir plus conditional, however, are exceptions to the Minimal Distance Principle; in example 2) above, «Lucy» could not correctly be interpreted as subject of the verb compraría. Echeverría anticipated that such structures would be difficult for children to understand and that discrimination between the two types of sentences would be associated with later stages of syntactic development (117). The Chilean children's comprehension scores indicated that decir plus subjunctive had already been acquired by the five-year old children, but that decir plus conditional was acquired later, showing a developmental pattern between the ages of six and ten (122-22). Echeverría concluded that children were overgeneralizing the Minimal Distance Principle and applying that strategy incorrectly with decir plus conditional (126).

Adjective Clauses. With regard to the use of relative or adjective clauses by Spanish speaking bilingual children, González (1971) noted that such clauses (e. g., «Yo tengo un monkey que baila» [I have a monkey that dances]) were used frequently by the Texas children at age 3.3 (76). González's frequency data for various age levels show relatively greater use of these clauses as ages of subjects increased, especially at age 4.6 and after (cf. 124, 138).

Brisk (1972) noted that relative clauses were «seldom used» by either the urban or the rural New Mexican five-year-old children, and that the only relative pronoun used in such cases was que. She observed that three of the four spontaneous examples in her data were sentences in which the «antecedent was missing», (e. g., «Mira que yo tengo» [Look at... that I have]). In some cases, Brisk noted, the children «simply juxtaposed» sentences (e. g., «La Julia tiene la baby [quien or la cual] todavía agarra la bottle» [Julie has the baby who is still holding/using the bottle]) (79-80).

Merino (1976) included as test items sentences representing two types of relative clauses, both of which involved co-referential noun phrases: 1) one in which both noun phrases function as subject (S-S) of their respective clauses

and 2) a second, in which both noun phrases function as object (0-0) within their respective clauses

(271).

Unfortunately, however, Merino's findings are reported only for the overall category of «Relatives» and not for each of these two types of relative clauses separately. In any case, regarding San Francisco children's comprehension of relatives, Merino reported a «developmental norm» over grade with children's performance improving with age; the effect, however, was not significant. Relative clauses, she noted, were among «the most difficult items» for the kindergarten and grade-two children to understand and among the easiest items for the older children (155). As for children's production

––––––––   492   ––––––––

of relatives was usually in second or third place across all grade levels (154).

McKay (1975) found that San Francisco children's production of adjective clauses did not differ significantly over grades one through four; adjective clauses, along with noun clauses, were used very frequently by all the children (85).

As for studies of monolingual children's Spanish, Gili Gaya (1972) observed only that the relative pronoun used almost exclusively by the first-grade Puerto Rican children was que. Fourth-grade children, however, showed increased use of quien [who] (e. g., «No hay quien salve a Pedro» [There isn't anyone who can save Pedro]) (92).

Echeverría (1975) investigated the development of relative clauses among Chilean children ages six to ten. He included two types of test sentences with different underlying structures. One, type A, was a sentence in which the subject of the dependent clause, as relative pronoun, was the same noun phrase as the subject of the main verb (e. g., «El niño que empujó a la niña cayó al suelo» [The boy who pushed the girl fell to the ground]). The second, type B, was one in which the subject of the dependent clause and the subject of the main clause were two different noun phrases (e. g., «El payaso que saludó Susana salió corriendo» [The clown (that) Susan greeted ran away]) (138). Echeverría found that comprehension of relative clauses was «a difficult task» for the Chilean children. Their identification of the subject of the main verb showed a developmental pattern, although it did not reach total accuracy. However, the children's accuracy in identifying correctly the subject of the relative clause depended on its position in surface structure. If the subject of the dependent clause, in the form of a relative pronoun, preceded the complement verb (type A), identification was at about 80-95%. If, however, the subject came after the complement verb (type B), accuracy was «extremely poor at all age stages», i. e., between 20% and 40% (142, 144). In terms of comprehension of the whole sentence, children's percent accuracy was low for both sentence types, although somewhat higher for A-type sentences than for B-type sentences across all ages. Regarding difficulty of comprehension of relative clauses, Echeverría concluded that his findings supported the result that type B sentences are more difficult to understand than type A sentences for children at any age level. Echeverría noted that even for A-type sentences, however, the comprehension level achieved was «far from denoting a command of these structures by the subjects interviewed» (147). He urged that further research test subjects older than age ten in order to determine children's level of comprehension of such structures at more advanced stages of language development.

Adverbial Clauses. Observations regarding the use of adverbial clauses by Spanish-speaking bilingual children are somewhat more frequent in the literature.

In his study of Texas children, González (1971) observed that locative clauses after donde were introduced at age 2.9; although use of locative clauses was more frequent among the Texas children after age 4.6, donde continued to be the only locative conjunction used (47, 123). Locative adverbials were introduced in the children's speech earlier than were temporal adverbial clauses. González observed that temporal clauses after cuando [when] were introduced at 3.3 years, but were used more frequently after age 4.0. Cuando was the only conjunction observed until age 4.0, after which antes que [before] and hasta que [until] were introduced among the 4.6 and 5.0 year olds respectively (123, 137). Causal adverbials after porque [because], although introduced at age 3.0, were of very low frequency. Conditional clauses after si [if] (e. g., «Si no se quita la camisa, le voy a pegar» [If he doesn't take off his shirt, I am going to hit him]), although introduced at age 3.0 (60), were used somewhat more frequently only after age 4.0. González reported no instances of si clauses introducing subjunctive among the Texas children at any age. Although he did not give explicit attention to the context of purpose clauses, González cited two examples which involved such clauses introduced by para que [in order that/so that]. One was a sentence used by one of the youngest subjects at age 2.6 (30). The second Gonzalez cited as an example of «grammatical deviation» involving failure to use subjunctive by a subject at age 3.6 (i. e.,«Démelo pa'que lo pongo aquí». [Give it to me so that I can put it here]) (94).

In her study of New Mexican five-year-old children, Brisk (1972) observed that adverbial clauses, although quite numerous, were introduced by only a few relaters, i. e., cuando,

––––––––   493   ––––––––

pa'que, (<para que) and si, the latter, only among the urban New Mexican children (80). Brisk, in her discussion of «compound» sentences, observed that the causal conjunction porque (e. g., «Ella no vino porque está mala» [She did not come because she is sick]) was used «by all of the children» (82).

McKay (1975), in contrast to the pattern of no growth for object noun and adjective clauses, found that San Francisco children's production of adverbial clauses increased greatly from first to fourth grade. Despite this apparent growth over grade, McKay noted that the effect of grade on children's performance was not statistically significant. The growth was more apparent, she noted, in temporal and causal adverbials than for locatives (52-53). Temporal adverbials were observed to be the most frequently used clause type at all but one of the four grade levels.

Merino (1976), in her study of San Francisco children's comprehension and production, included a complex sentence involving a purposive adverbial clause introduced by para que as one of three items in her «Subjunctive» category. As noted earlier, however, Merino's findings are reported only for the category of Subjunctive, and not for the separate semantic or syntactic contexts subsumed thereunder. No results are reported, then, for the adverbial para que context specifically. Merino also included a category called «Conditionals», which included complex sentences involving si clauses introducing feasible as well as contrary-to-fact events, and adverbial clauses introduced by al menos que [unless].¹⁴⁸ Again, however, Merino's findings are reported only in terms of the overall category of «Conditionals», which precludes any comparison of possible differences which may have prevailed in children's recognition or use of these various types of adverbial contexts. Regarding the San Francisco children's comprehension of Conditionals as an overall category, however, performance was «usually low and fairly stable» across all grades, peaking among the second-grade children and dropping off to about 50% accuracy among the fourth-grade children, which Merino attributed to language loss (174). In terms of order of difficulty, she reported that conditionals (and subjunctive) were generally «the most difficult categories» for children in all grades to comprehend (155). As for children's production of conditionals, Merino found significant differences across grade, with such differences usually favoring the older children. She noted, however, that even the children who performed best in this category, i. e., third grade, had «considerable difficulty» with conditionals, and that by the fourth grade, the bilingual children showed evidence of undergoing language loss (169). Regarding difficulty, Merino noted that for subjects of all grades, Conditionals were «the most difficult» structures to produce (154).

Regarding the language development of monolingual children, Gili Gaya's (1972) observations of Puerto Rican children's use of adverbial modification and subordinating conjunctions are of interest. He observed considerable use by the preschool children of the pa' que and a que conjunctions within purposive clauses (66). Causal porque was also observed frequently among the four-year olds, but not as frequently as para que. Gili Gaya noted that although the preschool children confused these two conjunctions, using them indiscriminately, six- or seven-year old Puerto Rican children distinguished purposive para que and causal porque (66). Within temporal clauses, certain subordinating conjunctions were used primarily in sentences of the youngest children, especially cuando and less frequently después que [after]. Among the fourth-grade children the inventory of conjunctions expanded to include antes que and tan pronto como [as soon as] (93). Use of the conjunction of manner, como [as], was observed among the school-age children. Conditional clauses after si were used by preschool children with subjunctive or indicative mood distinguished; however, as Gili Gaya noted, the frequency of such constructions greatly increased among the older Puerto Rican children (66, 93). He observed that even the four-year olds frequently used the coordinating conjunction pero [but]; similarly, he noted, aunque in the «adversative» sense of, pero (e. g., «Tenían buenas escopetas aunque no cazaron nada» [They had good guns, although/but they didn't get anything]) (137-38) was used frequently among the school-age children. Gili Gaya noted, however, that use of aunque in a concessive sense was rare. Noting the absence of concessive aunque among preschoolers and only an isolated case by a first-grade child (93), Gili Gaya suggested that concession presupposes «una complejidad mental» [a mental complexity] that young children have not yet acquired (138).

When we consider the finding of these various studies as they relate to use of subordinate clauses and complex sentences among Spanish-speaking children, it becomes clear that many factors operate to inhibit the formulation of generalizations at this preliminary stage in the research. Such factors include first, the paucity of systematic research regarding the topic and, second, questions related to methodologies employed in the existing literature which, in some cases, pose difficulty in interpreting the findings.

Many of the investigations to date, most of them doctoral dissertations, have undertaken exploratory or comprehensive descriptions of the syntactic development of Spanish speaking children. The nature of such studies necessarily precludes an in-depth and focused investigation of any one specific area of syntactic or semantic development. In several studies, use of subordination and complex sentences among Spanish-speaking children has been noted only as such structures provide the syntactic environment for use of subjunctive mood; in some cases, results are reported only for children's use of subjunctive, with no explicit attention to development of subordination or sentence complexity. The number of studies which have given explicit attention to the development of subordination among Spanish-speaking children remains very limited. Most of those studies which have observed use of subordinate clauses in complex sentences have noted children's spontaneous production of such structures. Study of children's comprehension of subordinate clauses has been very limited; the only environments to have received explicit attention in the literature to date are object noun clauses after decir, adverbial clauses of condition, and certain types of relative clauses.

In some cases methodological questions or problems have made difficult the interpretation of findings of particular studies; in such instances, the results reported could have been at least partly a function of the methodologies employed. Moreover, differences in methodological approach among studies which have treated children's subordination in complex sentences make comparison of findings across studies very difficult, if not impossible. In some studies the number of informants has been small; in such cases typically there has clear been very little quantification of data. All of the studies reviewed have been cross-sectional studies in which different children of specific age or grade levels served to represent various stages of linguistic development; none of the studies has been longitudinal. In some cases, especially in studies of bilingual informants, various intervening factors, e. g., degree of use of Spanish or language dominance, may have influenced the results. Many of the studies reviewed have observed only children's spontaneous use of subordinate clauses; rarely, however, has the use of such structures been elicited systematically by the investigator.

As the studies reviewed here relate specifically to the development of clause subordination and sentence complexity among Spanish-speaking children, they might well be characterized, then, as exploratory. Despite limitations regarding the state of the literature to date, however, some commonalities do emerge from the findings reported. It would be premature, of course, to formulate generalizations regarding this area of syntactic and semantic development among Spanish-speaking children on the basis of the limited literature to date. Nevertheless, examination of the results, however tentative, from past investigations can offer direction for future research into this most significant area of language development in Spanish, and merit attention within this context.

With respect to noun clauses, studies of bilingual (Brisk, 1972; González, 1971; McKay, 1975) and monolingual children (Gili Gaya, 1972) have observed early and frequent use of object noun clauses after verbs such as querer decir, and saber. Two studies of bilingual children have found that noun clauses which function either as subject or as predicate of the main clause are introduced later and are used less frequently than noun clauses functioning as object of the matrix sentence (Brisk, 1972; González, 1971). One study has shown that monolingual children's comprehension of object noun clause introduced by decir varies with the meaning of the verb and the underlying structure of the dependent clause: decir in a volitive sense introducing a subjunctive verb is understood earlier than decir in a reporting sense introducing a conditional verb. Comprehension of the latter construction apparently represents a more difficult task since it is an exception to the Minimal

––––––––   495   ––––––––

Distance Principle, which predicts the subject of the dependent clause (Echeverría, 1975).

Findings in those studies which have treated relative clauses among Spanish-speaking children seem at first glance to vary considerably. Fortunately, this is one area of clause subordination among Spanish-speaking children which has been studied from the point of view of both comprehension as well as production. Findings in studies of children's comprehension of relative clauses have shown some convergence. A study of bilingual children in San Francisco showed a developmental norm, with comprehension improving over grade (Merino, 1976). Nevertheless, both studies which have investigated subjects' comprehension of relative clauses have shown that such comprehension remains a difficult task even for older children, i. e., fourth-grade bilingual children (Merino, 1976) as well as ten year old monolingual children (Echeverría, 1975). A study of monolingual children has shown comprehension of relative clauses to be related to syntactic structure: adjective clauses in which the relative pronoun preceding the complement verb is subject of the dependent clause are more easily understood than are relative clauses in which the subject of the dependent clause does not precede the dependent verb (Echeverría, 1975). However, findings among those studies which have observed production of relative clauses, especially among bilingual children, are more divergent. Studies of bilingual children sometimes have reported low frequency (Brisk, 1972; Merino, 1976) and ill-formed structures (Brisk, 1972); other studies have reported early and frequent use among school-age children (McKay, 1975) and even among pre-school children (González, 1971).

With respect to adverbial clauses in the Spanish of children, investigators have typically observed their use in relation to semantic categories, such as temporal, locative, causal, and hypothetical. There is evidence that clauses of purpose after pa'que (<para que) appear early among monolingual (Gili Gaya, 1972) and bilingual children (Brisk, 1972). Regarding temporal clauses, studies of bilingual children have either suggested (González, 1971) or have shown (McKay, 1975) a develop mental trend. Evidence suggests that cuando is the first temporal conjunction used among both bilingual (Brisk, 1972; González, 1971) and monolingual children (Gili Gaya, 1972); investigators of both bilingual (González, 1971) and monolingual children (Gili Gaya, 1972) have observed that among older children the range of temporal conjunctions expands to include después que, antes que, hasta que, and tan pronto como. Evidence of developmental patterns for bilingual children's production of adverbials of cause (McKay, 1975) and of condition after al menos que and si (Merino, 1976) has also been offered. Locative clauses were observed to be introduced very early among Texas bilingual children; frequency of their use, however, increased only among the older children (González, 1971). In another study of bilingual children, production of locative clauses showed a pattern of growth over grade, although the pattern was not a very marked one (McKay, 1976). Other investigators of children's Spanish have made no mention of locative modification. The lack of mention of adverbials of concession in studies dealing with bilingual children and the specific mention of their rarity among even school-age monolingual children (Gili Gaya, 1972) suggests that concessive adverbials may have a later development than other types of adverbial modification. In other cases, however, attempts to discover commonalities in the findings of the literature to date are frustrated. For example, only one investigator has mentioned the confusion of causal porque and purposive para que among young monolingual children (Gili Gaya, 1972); such confusion has not been noted in other studies of Spanish-speaking children's use of complex sentences of which I am aware.

In general, then, an overview of the developmental literature as it relates to complex sentences and subordinate clauses in Spanish suggests that while children make developmental gains in this regard from the ages of two to ten, neither bilingual nor monolingual children at age ten have yet acquired the full range of semantic and syntactic expression characteristic of adult's use of subordination in complex sentences. Very few observations regarding Spanish-speaking children's use of dependent noun clauses have surfaced in the developmental literature to date. Even the limited literature on children's use of dependent noun clauses suggests that such structures develop variably according to the grammatical function they serve in relation to the main clause in complex sentences. Comprehension of relative clauses has been seen to

––––––––   496   ––––––––

be a difficult task for even ten-year old monolingual children, with difficulty related to the grammatical structure of the relative clause. The existing developmental literature suggests that children's use of adverbial clauses develops variably by semantic categories, with some types of adverbial modification appearing early, e. g., temporal, and other, e. g., concessive, not yet manifest in the Spanish of even fourth-grade monolingual children. The literature suggests, too, that the various semantic categories of adverbials have their own development, with the range of adverbial conjunctions in a given category expanding among older monolingual and bilingual children.

As for future research into this particular area of language development among Spanish speaking children, it goes without saying that the topic deserves to be the focus of concentrated and systematic investigation. While the early exploratory and comprehensive studies (e. g., Brisk, 1972; Gonzalez, 1971; Gili Gaya, 1972) were necessary and have contributed greatly toward our understanding of language development among Spanish-speaking children, the need exists now for in-depth and focused investigation of specific questions related to development of sentence complexity among Spanish-speaking children. In order to elucidate patterns in children's Spanish which are truly developmental, rather than dialectal, the norm of behavior against which children's language is measured must be the children's own language variety or dialect, rather than an external dialect, perhaps a standard dialect, to which the children have not been exposed. Systematic investigation of development of sentence complexity among monolingual Spanish-speaking children is of critical importance. Such studies would have the added benefit of providing a basis for comparison with investigations of bilingual children's language development, in which external factors may operate to mediate the developmental process. In order to identify features of bilingual children's Spanish which are truly developmental, efforts must be made to control for such factors, e. g., language dominance or degree of use of Spanish, which may otherwise confound the results. The effect of syntactic difficulty on children's development of subordinate clauses has been given limited consideration in the literature to date. Evidence from past studies suggests that this factor may influence the development of various types of noun and adjective clauses.

Efforts must also be made to determine to what extent semantic or cognitive development may also affect the development of subordination and sentence complexity in Spanish. The current literature for Spanish suggests that children's use of adverbial clauses is related to semantic development. To my knowledge, such questions have not been systematically addressed in the language acquisition literature for Spanish. Also, since subordinate clauses provide the syntactic environment in which subjunctive mood verb forms are used in Spanish, the development of subordination should also be studied in relation to children's morphological development, particularly, the morphology of the Spanish subjunctive. Attention should also be given to the possible relationship between development of sentence complexity and clause subordination in Spanish and corresponding development for other languages. The psycholinguistic literature includes a growing number of studies on the topic, especially as it relates to English.

Another interesting area of inquiry would be the relationship between first-language development and that of second-language acquisition regarding development of subordination and sentence complexity. Future studies of development of sentence complexity among Spanish speakers would also profit by greater attempts to relate empirical language data to theoretical issues and approaches to language acquisition. Fortunately, some work in this regard has begun for relative clauses in Spanish (Echeverría, 1975; Merino, 1976). The example which some of the studies have established, with larger numbers of informants and greater and more systematic quantification of data (e. g., Echeverría, 1975; McKay, 1975; Merino, 1976), is also one that might well be emulated in future studies of sentence complexity among Spanish-speaking children. Efforts such as these may well be expected to contribute significantly toward the growing literature on language acquisition in general and language acquisition in Spanish in particular.

Blake, Robert James. 1980. «The Acquisition of Mood Selection among Spanish-speaking Children: Ages 4 to 12». Unpublished doctoral dissertation, University of Texas at Austin. Ann Arbor, MI: University Micro films, No. 8100873.

Brisk, María Estela. 1972. «The Spanish Syntax of the Pre-school Spanish American: The Case of New Mexican Five Year Old Children». Unpublished doctoral dissertation, University of New Mexico. Ann Arbor, MI: University Micro films, No. 73-16585.

Chomsky, Carol. 1969. The Acquisition of Syntax in Children from 5 to 10. Cambridge, MA: M. I. T. Press.

Echeverría, Max Sergio. 1975. «Late Stages in the Acquisition of Spanish Syntax». Unpublished doctoral dissertation, University of Washington. Ann Arbor, MI: University Microfilms, No. 76-17457.

Gili Gaya, Samuel. 1972. Estudios de lenguaje infantil. Barcelona: Bibliograf.

González, Gustavo. 1971. «The Acquisition of Spanish Grammar by Native Spanish Speakers». Unpublished doctoral dissertation, University of Texas at Austin. Ann Arbor, MI: University Microfilms, No. 71-11540.

McKay, Maryann. 1975. «Spoken Spanish of Mexican American Children: A Monolingual and Bilingual School Program». Unpublished doctoral dissertation, Stanford University. Ann Arbor, MI: University Micro films, No. 75-13557.

Merino, Barbara Jean. 1976. «Language Acquisition in Bilingual Children: Aspects of Syntactic Development in English and Spanish by Chicano Children in Grades K-4». Unpublished doctoral dissertation, Stanford University. Ann Arbor, MI: University Microfilms, No. 77-7132.

Rosenbaum, Paul. 1967. The Grammar of English Predicate Complement Constructions. Cambridge, MA: M. I. T. Press.

Pedagogy

Teaching Elementary Schools

A Treasure Trove of Materials for the FLES Teacher

Folklore of Hispanic countries can provide the FLES teacher with a wealth of instructive and entertaining rhymes, jingles, and games. The integration of such activities can aid in the achievement of goals in listening, speaking, and cultural understanding when used in a meaningful communicative context. Alan Maley, talking about poetry and song as forms of language use, points out:

For example, the teacher who uses a little rima de sorteo to choose a leader for a classroom activity is providing input which is both comprehensible and culturally authentic. Although the children may not know the meaning of each and every word, they do know the purpose of the rhyme. (And, after all, how many native speakers of English know «Eenie, Meenie, Minie, Mo...» but who can define «eenie?») Furthermore, as MacRae reminded us in 1957, through hearing the language in larger units, of three or four lines rather than simply words, the learner can begin to imitate the intonations and melody of spoken Spanish (149).

By carefully choosing rhymes, games, and jingles which are related to the curricular con tent, the FLES teacher provides an easy way for students to become comfortable with the vocabulary. Soon they will be able to manipulate the familiar words much more efficiently and effectively than when taught by rote memory, drills, or simple repetition of lists.

Since many of the activities involve the body in some way, they take advantage, as Asher says, of that «powerful effect that the kinesthetic sensory channel has in making a second language learnable for most children» (Foreward).

In addition, the authenticity of the materials allows the children «to penetrate a new culture through meaningful experiences with cultural practices and cultural phenomena that are appropriate to their age level, their interests, and the classroom setting» (Curtain Anderson and Pesola, 150).

The purpose of this article is to offer a sampling of the many rhymes, jingles, and poems identified as part of the oral tradition. Included at the end of the article is a list of a few sources for the teacher who wishes to explore further this treasure trove of folklore.

Rimas de sorteo: (These little rhymes, said frequently and casually by the teacher, will soon be echoed by the children as they vie to choose leaders for their activities).

The most well-known counting-off rhyme is this one. It has many variations.

Rhymes and Jingles with Numbers: