Introduction
         Early 20th century American linguistics
         The role of language and culture in thought
A Structural Approach to Language
         Pre-Chomskian views
         Ladoís principles of language learning and teaching
         Behavioral theories of L2 learning and teaching
         A search for explanatory principles
Phonology
         Early sound production
         Vowels and consonants
         Phonology versus phonics
         First language interference
         Implications for teaching
         Graphemic issues in L2 phonology
Morphology
         Inflectional morphology
         Derivational morphology
         Learnability of English morphology
Syntax
         Word order
         Case marking
         Null elements and referential ties
         The role of syntax in construction of meaning
         Implications for teaching
Semantics
Discourse
         Multicultural issues
         Language socialization
Explanations for Second Language Acquisition
         First language effects
         arget Language Complexity
Language Development Stages
         Curricular adaptations
         Child first language acquisition
Issues in Reading
         Phonics, Phonology, and Orthography
         Vocabulary knowledge
 
 

         Second language learning research is rooted in the structuralist tradition of linguistics that has predominated in the United States since the turn of the century when anthropologists began studying native American languages.  The effects of that enterprise were to extend the boundaries of known characteristics of linguistic systems, and to posit a strong role for language and culture in the framing of thought and the formation of concepts within a society.  Such study also led to a more descriptive, as opposed to prescriptive, role for language study, and it eventually led to seeking explanatory principles for language learning as well.

Early 20th century American linguistics
         When anthropologists such as Franz Boas began studying native American languages, they were doing so in an effort to preserve the languages, since they knew that native peoples were disappearing.  In studying such languages as Navajo, Hopi, Eskimo, and Yawalmani (formerly spoken in Southern California), they discovered many interesting phenomena at the levels of sound systems, morphology, and syntax.  For example, in studying Yawalmani they discovered certain interesting patternings of vowels, so posited underlying vowel harmony rules.  In studying languages of the Southwest and Mexico, they found classifier systems such as those found in Chinese; in these languages objects are assigned classifiers before their names, usually associated with the shape of the object.  And in studying certain languages of Alaska and the Northwest, they found large numbers of words associated with snow and other weather phenomena.  As they learned such languages as Hopi, anthropological linguists like B. L. Whorf and Edward Sapir found that such concepts as noun and verb were problematic.  For example, when one talks about a house, one incorporates into the word house certain information such as whether it is empty or inhabited, whether it is in the process of being built, or whether it is on fire.  Given such a different grammatical structuring, these languages were seen by some anthropologists as causing people to have different cognitive structures as well.

The role of language and culture in thought
         Although subsequent work on human perceptual systems by Eleanor Rosch and others has strongly challenged the Whorf-Sapirian claim, it is nonetheless true that many people say that as they think and act in more than one language, so they conceptualize themselves and their world differently.  Some say it is the way their language is structured that causes them to do so; others say it is because they perceive the world through different cultural lenses.  Whatever the facts turn out to be, it is quite clear that boundaries among language, cognition, and culture are often "fuzzy."  However, for some purposes it is important to keep these concepts apart, and we will learn to think of linguists and anthropologists as professionals with overlapping, though different agendas and emphases.  Similarly, teachers also often associate themselves with a more linguistic or a more cultural view of the processes involved in acquiring a new language, fitting into a new classroom situation, and functioning as a member of two communities.  Again, in this course the goal is for inservice and preservice teachers to understand principles of culture and language as interrelated, but also separable, not only for study, but as they apply in classrooms.

Back to Table of Contents

Pre-Chomskian views
         It was stated above that American Structuralism in anthropology and linguistics pervades the way linguists in this country think about language.  A main assumption about the nature of language is that it has several somewhat independent formal levels, or components: phonology, morphology, syntax, the lexicon (words) and semantics, and discourse.  Anthropological linguists through the '50s had concentrated mostly on phonology and morphology until the generative-transformational challenge by Noam Chomsky in the late '50s.  While the study of language learning was strongly influenced by behavioral psychology from the '20s through the '50s, it has become more allied with Chomskian views since the early '60s.  Nonetheless, the pursuit of understanding the structure of language in the mind dominates the thinking of many applied linguists studying the processing and learning of languages.  The formal structures that Chomsky seeks to explain are phonology, morphology and syntax.  Semantics and discourse are those components that interact with other cognitive mechanisms, those associated with meaning and the socially determined aspects of language behavior.

Ladoís principles of language learning and teaching
         Lado, R.  (1964).  Language teaching A scientific approach. NY:McGraw Hill.  Chapter 4 - A Modern Theory of Language Learning

        ìTo formulate a theory of second language learning we must know what is involved in the process of speaking and listening, so that we may account for learning that process.î (p. 32).  Note that the explanation for learning was to be found in behavior.  At this time psychology was the scientific study of behavior, not of the mind.  Apart from the mechanism of stimulus-response, it was assumed that what actually goes on in the mind could not be known.  At the time Lado published his book such concepts as mental models, schemas, and information processing mechanisms had not been developed.  Nor had the notion of generative grammar and its corollary concepts of deep and surface structure even been introduced.  Although Chomsky had challenged Skinnerís theory in 1959, his own Aspects of a Theory of Syntax did not appear until 1965.

         The process of speaking and listening is described as follows: ìThrough some motivation the person decides to speak, and some content is brought under attention.  Through association of this content with expression in the language, sentences are constructed with words, intonation, phonemes, etc.î  (p. 32).  This rapid process requires facility and adequate memory to maintain the normal very rapid pace of production.  The process of listening is a reversal of the process of speaking.  ìReading and writing are parallel to listening and speaking.î (p. 33).  Thus, inherent in production and comprehension are the following: attention and facility, recall and memory span, motivation and will, and fluency and motor function.

         At the time of Lado, research on learning had been of the following kinds: conditioning, verbal learning, and trial and error.  Lado found these types to be inadequate.  ìThe fact is that language learning research to be fully relevant must deal with language structure in use or with tasks which are clearly relatable to language structure in use.î  (p. 36).  His basic underlying model entails three aspects:

        Units and patterns of expression <- association -> Units and patterns of content

        Research on how form encodes meaning ìmust encompass the simplest phoneme as well as the most abstract cultural or individual value.î (p. 36).  Note that Ladoís learning theory would treat all phenomena to be learned in foreign language study as of the same type.  You will learn that Chomsky more narrowly circumscribed the foci of language study (phonology, morphology, syntax, and lexical structure).

         Following are some of the laws of learning that have been proposed in the general learning literature (still driving instructional design approaches), although Lado questions whether they might hold for language learning:
         Law of contiguity - things learned together will be associated in recall
         Law of exercise - practice makes improvement and enhances recall
         Law of intensity - intensity of practice enhances learning
         Law of assimilation - new stimulating conditions elicit same responses as earlier similar     conditions
         Law of effect - responses associated with a satisfying state of affairs are reinforced;     responses associated with an annoying state of affairs are avoided

Behavioral theories of L2 learning and teaching
 ìLearning a second language is defined as acquiring the ability to use its structure within a general vocabulary under essentially the conditions of normal communication among native speakers at conversational speed.î  (p. 38).  Here Lado is defining the parameters of language learning.  Note that his definition is again behavioral; it is almost not a definition at all, and it certainly is not an explanation.  His principles are somewhat more helpful:

Robert Lado's principles of language learning
ï  Expression, content, and their association are learned in no sequence and are independent from each other
ï  Learning associations - need to learn expression with content and content with expression
ï  Learning through experience - (exception: analogies with earlier learned knowledge)
ï  Trace, habit, facility - Experiences leave traces; habits are developed through tracing; facility comes about when habits are formed
ï  Prior experiences (that are repeated) contribute to language learning
ï  Facilitation and interference - when expression, content, and association are functionally the same in L1, L2 learning is facilitated; when they differ, there is interference
ï  Native language facilities may be activated first
ï  Total and partial experience and set - sometimes one learns in wholes, sometimes in parts; breaking things down for mastery can be helpful; silent rehearsal can help
ï  Attention and awareness - attention must be freed from smaller units to larger ones; some functions at one level must become automatized
ï  Volition and learning - will controls attention and practice
ï  Motivation - urge to communicate increases language learning
ï  Memory span and motor function - learning continues through exercise after a pattern becomes (at least partially) available; experiences increase memory span; ability develops to note errors
ï  Availability of response and recall time - learning is manifested by decreased time for earlier-learned items to be retrieved from memory
ï  Individual differences - different rates of learning for different learners under similar conditions (aptitude differences)
ï  Order of skills - listening and speaking are ìtotal language experiencesî while reading and writing are ìpartialî ones.  Partial skills easier to learn after total skills are learned, but not the other way around.

Robert Lado's principles of language teaching
ï  Introduce speech before writing
ï  Have students memorize basic sentences - shorter memory span in TL than in L1; therefore dialogs help students memorize language in context
ï  Have students practice patterns to develop habits (then use analogy, substitutions, variations, transformations) - donít talk about the TL; use it
ï  Teach sound system for use - use props, articulation guides to help
ï  Control vocabulary presentation - maximize teaching of sentence structures first; substitute vocabulary when patterns learned
ï  Highlight dissimilarities in L1 and L2 - provide the most practice in these areas
ï  Maintain distinctions between spoken and written language; written language is often an inadequate representation of speech (closest - Spanish, not so close - English; totally unrelated - Chinese)
ï  Teach patterns gradually, in graded steps, simple to complex
 a.  Teach sentences as beginning point, not words
 b.  Introduce subsentence elements in connection with sentence frames
 c.  Add new elements to previous ones - teach yes/no questions with do before teaching interrog wds.  d.  Adapt learning difficulty to studentsí capacities e.  Keep dialogs natural, but try not to add too many new patterns
ï  Provide opportunities for practice, not translation.  Translation is a far more difficult process than the four skills.  Seeking word for word translation of materials gets in the way of studentsí understanding that each language is structured differently, and there are no exact word for word parallels
ï  Teach authentic language standards, language as it is, not prescriptive uses
ï  Practice - quality and quantity of learning depends on amount of practice
ï  Shape responses - through partial experiences and props (like backward build-up, a diagram of the mouth, etc.) help student to do the sentence accurately
ï  Maintain naturalness in speed and style
ï  Immediate reinforcement - corollary to 12 - praise successes right away
ï  Impart positive attitudes toward target culture
ï  Teach content through its cultural milieu
ï  ìTeach primarily to produce learning rather than to please or entertain.î

A search for explanatory principles
         When Noam Chomsky challenged the premises of Skinner's work, laid out in Verbal Behavior in the late '50s, it was with a set of structuralist arguments.  Although Skinner had claimed that language learning, like all other kinds of learning, was simply a matter of stimulus-response mechanisms acting on linguistic input, Chomsky claimed a whole other set of mechanisms for language acquisition, built on a whole other set of assumptions about the nature of language.  His theory of language assumed first that language structure is innate, creative (or generative), modular, species-specific (human only), and principled.   That is to say, the child is born with a structure that predisposes him or her to take incoming linguistic data and organize it according to predetermined principles that only humans are born with.  These principles are specific to language, and exist in a separate part of the mind, the language module or faculty.

         Chomsky's position is supported by even a casual observation of children.  First, although they occasionally repeat what others say, children also make up countless numbers of original sentences.  Second, the "mother tongue" seems not to be learned through formal study, but is acquired by all children by about the age of five simply through interaction with other members of the community.  Language learning being such an amazing feat, a cognitive structure with considerable subtlety and sophistication must be in place earlier than other cognitive systems, many of which require more maturity and/or formal schooling experiences in order to be activated.  Third, even though apes have been taught to use sign language and computers to communicate, their communication systems seem to lack the hierarchical structure that human language entails; human languages appear to be hierarchically, as opposed to linearly, organized.  For example, on the surface the following two sentences have the same structure, but they are not structurally parallel:

 1. John is eager to please.
 2. John is easy to please.

In 1) John is the subject of both is and please, while in 2) John is the subject of is, but the object of please.  It is doubtful that non-humans will ever be found to have communication systems that can parallel this feature.  Although we will not have time to study Chomsky in detail, his influence in linguistics and language acquisition is very great, and his ideas of innate language structure are very important.

Back to Table of Contents

Early sound production
         Phonology is the study of the sound system of a language.  It appears that the sound is the first part of the total system that babies discover.  Babies begin experimenting with the phonological system of the language they hear around them in the crib.  As they babble, they begin to sound more and more like they are speaking even though their sounds are unintelligible; they learn the tune before the words.  It seems that although the brain is predisposed to organize sounds into a language, if the sounds of a language do not stimulate the portion of the brain prepared to receive them, then that section of the brain does not develop.  Preliminary research indicates that if the nerve cells predisposed for speech are deprived of stimulation from the sound of speech, they are used by the brain for another function.  As children we gradually become experts in reproducing the sounds of our own language, but we do not learn to produce sounds we do not hear.  For example, as adult English speakers, we will find it hard to reproduce the initial 'ng' of Vietnamese or the 'rr' of Spanish.

Vowels and consonants
         There are two basic kinds of sounds.  Vowels are sounds that leave the vocal cavity unobstructed, and are shaped by various positions of the tongue.  Consonants are sounds that obstruct the flow of air through the vocal passage with contact points of the tongue being in different points along the roof and teeth.  The lips are also used to form some vowel and consonant sounds made toward the front of the mouth.  In English some consonants are produced with voicing and others without.  These occur in pairs.

 Activity: Produce the following consonant sounds and notice the manner and point of articulation:  the 'p' in 'pot,' the 'sh' in 'shine,' the 'd' in 'dinner,' the 'l' in 'light,' the 'g' in 'game,' the 'v' in 'vase,' the 's' in 'sit,' and the 'b' in 'bed.'  Explain similarities between 's' and 'sh', 'p' and 'b', and 'b' and 'v'.
 Produce the following vowel sounds and notice the position and shape of your tongue and lips: the 'a' in 'father,' the 'e' in 'scene,' the 'ow' in 'now,' the 'i' in 'pinch,' the 'u' in 'put,' and the 'oo' in 'look.'   Make a chart in the shape of a tic tac toe diagram of where you think you produced the various vowel sounds.

         Each language has a unique sound system, and while languages have similar sounds (called phonemes), they are nonetheless distinct.  When we speak of the sounds within the system of a particular language, we refer to them as phonemes and represent them as letters within slashes.  When we speak of a speech sound in isolation from a particular linguistic system and wish to characterize its articulatory features more preceisely, we refer to them as phones.  For example, both English and Spanish have /t/, but since they are produced with slightly different places or manners of articulation, they are phonetically distinct.  That is, with sharp ears, or with spectrographic analysis, one can detect a difference.  The English /t/ in word-inital position is produced with the tip of the tongue in contact with the alveolar ridge (right above the upper teeth in the middle), and it is aspirated (accompanied by a puff of air).  The Spanish /t/ is produced slightly further back, on the blade, or flat part, of the tongue against the teeth; it is also unaspirated.  A /t/ sound in a language such as Vietnamese that is produced closer to the bottom of the upper teeth more closely approximates the English /th/ sound, the one you hear in think and three.  One that contacts the roof of the mouth further up, such as in some languages of South Asia, more closely approximates the English /d/ sound.  Consonant sounds may also differ in the flow of air on which they travel.  For example, speakers of languages like Amharic, spoken in Ethiopia, produce some sounds on the intake, rather than on the expulsion of air.

 Activity:  With a person whose first language is not English and who speaks with an ëaccent,í identify some consonant sounds that s/he makes that sound 'foreign' to you, and ask him or  her to help you learn to make them.  Note how the sounds are similar to consonants in English, and how they are different in terms of place and manner of articulation, voicing, and air flow.

Phonology versus phonics
         English has approximately 42-44 different sounds, including about 13 vowel sounds.  Obviously then, there are more sounds, or phonemes, in English than there are letters in the English alphabet.  The fact that the sound-symbol correspondence in English writing is rather poor, especially in many high frequency words, makes it a difficult language for many non-native speakers to learn to read and write, especially younger children who may just be beginning to read and write their own languages.   The technique for teaching reading called Phonics is an attempt to deal with the regularities and some irregularities of the English sound-symbol correspondence.  However, it should not be assumed that phonics instruction is based on a study of phonology, a science that is much more precise and universally applicable than phonics.  How a phonics approach to teaching a person to read a second language can be extremely confusing will be the subject of a later lesson.
 Activity:  To give you an example of the difficulty faced by learners of a language that has a different phonological system than English, consider the word 'fish,' and see why 'ghoti' could be a spelling of it, given the rules of English spelling.

First language interference
         By the time we are in elementary school, our brains have processed the phonological system, or the tune, of our language, but as we grow older, particularly beginning with the time of puberty and beyond, significant problems can and do occur when one meets the phonemes of a new language, difficulties quite unrelated to the system used for representing the sounds with letters of the alphabet.  For example, some languages do not distinguish the vowel sounds in words like 'sit' and 'seat' (Spanish, Slavic languages) or the initial consonant sounds in 'lift' and 'rift' (East Asian languages).  Much of the problem one has with learning the English sound system can be attributed to interference or transfer from the native language.  Sometimes the learner can hear the sound, but be unable to produce it; sometimes s/he may not be able to hear a difference at all.  For example, English speakers learning Spanish may have great difficulty with the /r/ phonemes in that language, and may not be able to produce a difference between the words 'pero' and 'perro,' even though they may be able to hear the difference.

Implications for teaching
         In general, giving an LEP student practice with the sound system of English need not be the responsibility of the content teacher.  However, since phonological problems are bound to occur, the teacher should be able to identify when phonological problems interfere with comprehension or effective participation by the student, devise activities that will help overcome those problems, and assign peer tutors to assist.  For example, speakers of Slavic languages could be given targeted assistance with content words having diphthongs (also called 'long vowel sounds') to pronounce, since Slavic languages do not have them.  Arabic speakers could practice words that begin with /p/ and /b/ to help them distinguish those consonants more clearly.  Students may be motivated to modify their pronunciation toward more target-like norms when they realize that people do not understand them.  However, pronunciation can not improve in the absence of opportunities to speak, and practice with native speaking peers and teachers is essential.

Graphemic issues in L2 phonology
         One of the most difficult aspects of English pronunciation that has a direct impact on the teaching of reading is the English system of vowels.   This is particularly important when we consider the vowel systems of other languages.

Spanish has 5:  a (father) e (red)  i (beach) o (phone) u (tube)

Vietnamese has 11 simple vowels and 6 clusters (diphthongs?); each syllable of the (mostly) mono-syllabic language is assigned one of 4 (So. & Central VN) or 5 (No. VN) tones.

English has 9 simple vowels with /y/, /w/, or /H/ (the sound in /r/-less dialects) possible with each, making a possible 36 combinations, all of which occur in at least some dialect of English.

Each dialect of Arabic has slightly different vowel sounds, but there are generally only about 3 of them, /u/, /iy/, and /a/.  These vowels are not usually written, but simply appear as diacritical marks on the consonant of the syllable to which they are attached.

Question:  Given the above facts, what difficulties or advantages would one of these groups of speakers have in learning to distinguish among and pronounce English vowel sounds?  Does regional variation in American dialects (either of the teacher or the students) create any problems for children learning to read?  If so, how?

Pronunciation of letters:  Many languages of the world having a five vowel system (or fewer) and the Roman alphabet assign the letter 'a' to the pronunciation of the sound /a/ in 'father,' the letter 'e' to  the pronunciation of the sound /e/ in 'bed' or /ey/ in 'bake,' and the letter 'i' to the sound /i/ in 'pick' or /iy/ in 'peek.'  However, in our pronunciation of the Roman alphabet the pronunciation of our letter 'a' is similar to /e/ or /ey/; our 'e' is similar to /i/ or /iy/, and our 'i' is like /ay/ in the five-vowel languages.

Question:  How does this set of facts help you understand some of the reading and/or spelling problems of second language students?  Can you think of examples from your own experiences?

        Teaching techniques demonstration:  Effective teaching of sound differences usually sequences auditory discrimination (hearing the difference between two sounds) activities prior to production activities.  Production can be aided by such techniques as imitating the sound using hands, drawing parts of the inside of a mouth, using a mirror, or watching the teacher produce the sound.  (In some cultures, drawing attention to the mouth, especially the tongue, can be rude and embarrassing, so teachers are advised to circumvent the problem.)  Steps one might use to develop auditory awareness of vowel or consonant contrasts are the following:

 ï  have student(s) listen while you say words in minimal pairs several times.  Label one word '1' and the other word '2.'  ((1) bag, bag, bag, (2) back, back, back - bag, back; bag, back, bag, back)
 ï  have student(s) hold up one finger if you say 'bag,' two if you say 'back'
 ï  do this until s/he/they can identify the two sounds with no difficulty
 ï  add a third word and label it '3' (ban).  This complicates the task slightly and forces learners to be more attentive.
 ï  have student(s) say the words as you show 1, 2, or 3 fingers.  Assist with drawings etc. as necessary.

Drills for speakers of various languages
The following pairs of words present problems for the L1 Spanish speaking children so might be helpful for practice:  bag - back, pig - pick, leaf - leave, cap - cab, fat - fad, cud - cut, thin - then.  Spanish speakers also require practice with the i/iy (pick/peek) and e/ey (pen/pain) vowel contrasts.

The following word pairs would be useful for helping learners who speak a variety of Asian languages to discriminate and practice:  the pairs above as well as gas - gash, laid - lathe, pal - pan, rake - lake, words ending in /r/ or any fricative, or any consonant cluster!

The following pairs would be useful for speakers of Slavic languages and Arabic, for they have fewer tense vowels than English does:  bait - bet, man - men, meat - mit

Back to Table of Contents

         In the previous section you were introduced to the idea of a phonological system, a system of discrete sounds and patterns of sound relationships in a language.  The sounds of a language are not in themselves associated with meaning as morphemes are.  Morphology is the study of the smallest units of meaning of a language and the processes by which words become created and modified.  They are of two general types: free and bound.  Free morphemes are words such as 'cat' to which are assigned referential meaning.  Bound morphemes are 'pieces of words' to which meaning can be assigned, but which must accompany a stem to form a unit, such as the plural '-s' on 'cats.'

         Morphology is not the study of meaning per se, but of the attached or otherwise associated bound forms.  Since it is a study of form rather than of referential meaning, some linguists consider morphology to be a part of the syntactic system of a language.  In some instances it is also linked to the study of phonology.  Depending on their position in words, and with respect to the phonemes adjacent to them, morphemes can take different forms.  For example, although the regular English plural is written as '-s' or '-es,' it is pronounced /s/ only after voiceless obstruents (cats, ducks, lips, laughs).  After /j/, /zh/, /sh/, /z/, and /s/, it is pronounced as /+z/ (judges, kisses), and elsewhere it is pronounced as /z/ (dads, bugs, hams, laws).  Another typology of morphemes is the distinction made between bound morphemes that either modify words or create new ones: inflectional and derivational morphemes respectively.  Note: in this material morpheme boundaries are indicated with a hyphen either before or after the affix showing where it must be attached to a root word.

Inflectional morphology
         Inflectional morphemes are those that "modify a word's form in order to mark the grammatical subclass to which it belongs" (O'Grady, Dobrovolsky, & Aronoff, 1993, p. 130).  English really has only a few inflectional morphemes, as the following list illustrates:

 plural                     -s
 possessive             -'s
 third person sglr present   -s
 progressive           -ing
 past tense             -ed
 past participle       -ed, -en
 comparative           -er
 superlative             -est

         In order to understand the importance of the acquisition of an inflectional morphological system we need to make a small comparative study of several languages, some of which are spoken by children in our state.  We have already seen that English has a rather simple way of marking nouns, with either a plural or singular form.  However, Spanish, Italian, and other Romance languages also gender-mark nouns.  For example, in Spanish 'table' is 'mesa,' a feminine form, and 'book' is 'libro,' a masculine form.  Articles preceding and adjectives following these noun forms must also agree in gender and number with the nouns, as in the following examples.  Note also that verb suffixes, '-a' and '-an' agree with the singular and plural nouns respectively.

 3. El libro rojo está en la mesa pequeña.
  The book red is on the table small
 4. Los libros rojos están en las mesas pequeñas.
  The books red are on the tables small

        Other kinds of noun marking also occur in languages.  Many indigenous languages spoken in North and South America use classifier systems according to their shape.  Shapes often marked in classifier languages include 'round' and 'long, cylindrical'.  A word like 'sun' would thus include two parts, the classifier and the name of the thing itself.  Vietnamese and Chinese also have classifier systems, though the systems include semantic information as well as information about shapes.  For example, in Vietnamese 'cat' is 'con meo' and 'pig' is 'con heo,' 'con' being the classifier used with the names of certain animals.  Other languages such as the Bantu languages in Eastern and Southern Africa have concordance systems.  For example, Swahili has eight noun classes, most having singular and plural forms.  These are basically semantic classes, such as the 'm-/wa-' class that includes most of the names of animals and humans, or the 'ki/vi' class that includes the names of many small objects.  When a language has a noun concordance system, adjectives, possessives, locatives and verbs all carry the concordance markers of the head noun, as in (5).

 5. Watu wale wawili walikwenda sokoni.
  People those two  3ps-past-go  market-to
  "Those two people went to the market."

        Finally, many Europeans, specifically speakers of Slavic, Germanic, and Baltic languages, case-mark nouns.  In Czech, for example, /muzh/ is nominative (used as the subject), /muzhe/ is genetive (possessive), /muzhi/ is dative (indirect object), and /muzhe/ is accusative (direct object).  In German the article is used before the noun marks it for case, but in some in some cases also marks the noun itself: der Mann (nominative), den Mann (accusative), dem Mann (dative), but des Mannes (genetive).  Some languages have as many as seven or eight cases.  English carries a remnant of case marking, but only in its pronoun system (they, them, their, themselves).

 Activity in class:  Participants observe and comment upon the following system of objective, possessive, and reflexive pronouns in English, and tell ways in which they are consistent/inconsistent and redundant/non-redundant.  They should focus particularly on the 3ps masculine paradigm, then discuss whether 'standard English' is always logical or regular.

         Objective: me, you, him, her, it, us, you, them
         Possessive: my, your, his, her, its, our, your, their
         Reflexive: myself, yourself, himself, herself, itself, ourselves, yourselves, themselves

 Derivational morphology
         Derivation is the process by which a new word is formed from a base word, such as when we add 'un-' to 'happy' to form 'unhappy.'  Sometimes derivational morphemes are category-changing, such as when we form 'kindness,' a noun, from 'kind,' an adjective, and the suffix '-ness.'  English has quite a rich array of derivational morphemes that speakers add to words to augment their meanings.  These are called affixes; prefixes precede the base while suffixes follow it.  Following is a short list of common affixes, taken from O'Grady et al (1993, p. 121).  In reading these you may notice that some can have more than one meaning:

 Affix               Change          Semantic effect         Examples

 (Suffixes)
 -able               V -> A          able to be X'ed          fixable
 -er                  V -> N          one who X's              worker
 -al                  N -> A          pertaining to              national
 -ize                 N -> V          put in X                    hospitalize
 -ate                 A -> V          make X                     activate

 (Prefixes)
 ex-                  N -> N          former; out              ex-president; excerpt
 in-                  A -> A          not X                       incompetent
 re-                  V -> V          X again                     rethink

        In addition to these affixes we have a great number of Latin and Greek roots and affixes that we use to build our technical lexicon, an example being 'mononucleosis' - 'mono-' (one), 'nucle-' (nut), '-osis' (condition).   Although sometimes knowing the meanings of roots and affixes that come from other languages will help us understand words better, the fact that 'mononucleosis' is "an acute, infectious disease" (American Heritage Dictionary, 1982, p. 811) is not quite so obvious.  This is because as time goes on speakers expand and change word meanings, sometimes through metaphorical extension, as in the case of 'nucleus' (nut), that "central or essential part around which other parts are grouped or collected" (p. 851).  More about how word meaning is assigned is presented in the section on semantics.

         While affixing is one way speakers create new words, compounding of words such as 'cowboy' is another.  Other word formation processes include: clipping - 'prof' from 'professor,' acronyms - 'radar' for 'radio detecting and ranging,' blends - 'brunch' for 'breakfast' and 'lunch,' and backformation - 'enthuse' from 'enthusiasm.'   Rather than being built by an additive process as in affixing, back-formation is subtractive.  Speakers assume that words made of several common morphemes have been built up by affixation.  For example, long ago speakers made up the word 'peddle' from 'peddler,' by analogy with 'worker' - 'one who works,' even though 'peddle' as a verb did not occur in the language before.  This is a common psycholinguistic process called overgeneralization, and it occurs frequently in learner data.  Second language learners, like young first language learners, often overgeneralize, or extend, the meaning of a word to cover more semantic territory than it actually covers.   More on this issue will be discussed in a later section.

 Activity:  Select fifteen words from your field of study that contain Latin or Greek roots,  prefixes, or suffixes and that your students will be expected to know.  For each of the words, tell us whether a knowledge of the core meaning of the word can be figured out by an analysis of its constituent parts:  scientist (scien + -ist  - OK); define (de-+  fine - NO!)

 Learnability of English morphology
         In the preceding section on inflections we have seen that compared to other languages English has a relatively simple system of marking nouns.  If we were to study verbs and other lexical categories we would find similar complexity in systems different from English.  Therefore, for native speakers of those languages, the English system of plural marking should be easy (especially for speakers of languages in the South Pacific that have forms showing a three way distinction: singular, two only, and more than two).  However, some languages spoken in Southeast Asia, of which Vietnamese is the best example, have no affixation whatsoever on any word.  Thus, plurality can only be shown by specifying number, or using a word like 'many.'  When speakers of Chinese, Vietnamese, and Thai begin learning English they have difficulty with the plural endings, as well as the verb tenses (final '-s' and '-d' tend to create complex final consonant clusters that speakers of Asian languages find difficult to pronounce).  Thus, acquisition of the English plural and past tense forms is often problematic for the speakers of languages such as Vietnamese; these forms may not occur regularly, even in the speech of well-advanced learners.  While omitting plurals may not seriously interfere with a native English speaker's comprehension of such learners, their failure to produce past tense inflections sometimes causes listeners difficulty regarding when events being refered to actually occurred.

         We can not leave this section with a misperception about the nature of the English inflectional morphology system.  Because it lacks the large number of forms that other languages have, one might think it is easy to learn.  But it is with these very simple morphological forms that non-native speakers experience the most difficulty, and children learning English as their first language also learn some parts of the system late.  Both Slobin (1979) and Jaeggli and Hyams (1987) have suggested that morphologically complex, but regular, systems are easier for small children to learn than reduced or irregular ones.  Children with Italian or Spanish as their first language learn the regular paradigm for verb-marking (i. e., Spanish '-ar' verbs such as 'hablar': -o, -as, a, amos, aís, an) earlier than English-speaking children learn to mark third person singular '-s' on the verb (I play, we play, she plays).  Systems in which the same morpheme is used for more than one purpose are also difficult for learners.  For example, in English 'do' is both a semantically empty function word and a verb in its own right, in Czech the form /muzhe/ is used for more than one case, and in the German genetive both the article and the noun must be case-marked.  Areas of any system where there is difficulty for first language acquirers are problematic for second language learners as well.

 Class activity: Participants construct charts showing the various components of the English verb system.  From these charts they can then try to predict what may be difficult for their students.

Some possible components include the following:
         ï the behavior of English modals and auxiliaries in questions, negation, emphasis, contractions, and tags
         ï the English modals (should, can, etc.) and all the ways in which they are used
         ï ways in which the 'be' forms (is, are, was, etc.) can be used (with V + ing and V + ed/en)
         ï regular and irregular verb forms (look for systematicity in irregular forms)
         ï the forms of 'do' and when 'do' support is used and not used
         ï the occurrence of the English past participle: with perfective aspect and passive voice
         ï spelling versus pronunciation of English past tense morphemes ('-ed' vs /t/, /d/, /+d/)

Back to Table of Contents

Word order
         Syntax is the word used to describe relationships of constituents to one another in sentences of a language.   What appears to be given in all languages is that words can not be randomly ordered in sentences, but that they must be arranged in certain ways, both globally and locally.  For example, in English the normal way of ordering elements is subject, verb, object (Boy meets girl).  Subjects and objects are composed of noun phrases, and within each noun phrase are elements such as articles, adjectives, and relative clauses associated with the nouns that head the phrase (the tall woman who is wearing a hat).  Native speakers of a language seem to have a sense about the order of constituents of a phrase, and such knowledge appears to be outside of what one learns in school.

 Activity:  Arrange the following words into a properly ordered noun phrase (not a sentence).  Check to see if you can arrange the words any other way and still have a natural-sounding phrase.

                     two      cereal      large      those      bowls      yellow

 Activity: Can you come up with a rule about the order of adjectives before nouns in English?  If you know another language, translate the words and arrange them properly according to the rules of that language.

         There seem to be general patterns across the world's languages in terms of word order.  Within the last 20 years theoretical linguists have identified two basic orders: XC and CX, head (X) first and head last respectively.   To make this clear we begin with the idea that there are four kinds of phrase, or XPs, X representing one of the following categories: nouns, verb, adjectives, and prepositions.  An XP (i.e. noun phrase, verb phrase) has two elements: a head and a complement.  In each of the following phrases we can see that English follows the XC word order (heads are underlined):

 6. NP: the man with a golden arm
 7. VP: polished the furniture
 8. AP: angry as a hornet
 9. PP: with a song in my heart

Case marking
         In some other languages of the world such as Japanese and Korean, however, the typical order is CX.  This means that objects precede verbs, and relative clauses precede nouns.  Languages can violate these orders, however, when other kinds of information is provided, such as case-marking on nouns.  Thus, in Japanese, an SOV (subject, object, verb) language, one might say 'horse dog kicked' and mean that the dog kicked the horse (As for the horse, the dog kicked it), provided that the two nouns are properly case-marked for topic and object.  Although English, like German and other Germanic languages, once had noun case marking and freer word order, it now adheres rather strictly to the canonical, or usual, SVO order.   Objects can become subjects in passive sentences only with a strongly marked verb form (be + V + -ed/-en: i. e., was given, is mended, are completed) that makes it clear that the sentence subject is not the agent of the action.  However, because the difference between the English participial forms (-ing and -ed/-en) and their functional distribution may not be clear in the minds of LEP learners, passive sentences may be understood as active ones.

Null elements and referential ties
         Word order and case marking are not the only phenomena treated in the study of syntax.  Two others deserve mention here.  One pertains to whether or not languages permit certain elements to be null, or missing.  In such widely different languages as Spanish, Italian, Vietnamese, and Thai subject NPs can be omitted from sentences.  For example, in Spanish it is permissible to say the following (10), whereas in English the deletion is unacceptable (11):

 10. Habla inglés.
  Speak-3ps English
  He speaks English
 11. *Speaks English.

In Spanish and Italian a verb ending (in this case the '-a' attached to the verb stem 'habl-') provides the needed information, for '-a' uniquely identifies the subject as third person singular.  In Vietnamese, however, no morphological marking allows us to retrieve the deleted subject.  Instead, the topic of the discourse seems to be carried across sentence boundaries until a new subject is needed.  In other languages such as Farsi (spoken in Iran) certain positions that may be null in English must be filled.  Sentence (12) is an example of the use of a resumptive pronoun, often found in the writing and speaking of Iranian learners of English

 12.   *The man that you saw him was my father.

         A fourth characteristic of languages has to do with internal and external referents of pronouns.  For example, in (13) 'herself' can refer only to Farah, because a referential tie between an NP and its reflexive counterpart can not hold across non-finite clause boundaries in English.   However, ties across clauses may occur with object pronouns, as in (14).

 13.   Mary i told Farah j to help herself j.
 14.   Mary i told Farah j to help her i,k

Note that in (17) 'her' can refer to 'Mary' or some other person not mentioned in the sentence.  However, it can not refer to 'Farah,' as it is permitted to do in Chinese, for example.  Thus, in languages of the world referential ties may be made over a range of constituent structure types, and may also occur beyond the level of the sentence.   The different rules of NP deletion and differences in referential ties across a discourse may add another layer of difficulty to the comprehensibility of L2 speakers as they interact with native English speakers.

The role of syntax in construction of meaning
         It should be obvious that a knowledge of syntax (be it tacit or explicit) is a part of the general knowledge of language that underlies the way we produce and understand the spoken and printed word.  Sentence elements are not randomly ordered constituents, and the word orders of languages differ in principled ways from each other, even though there seem to be underlying general patterns.  Some languages mark nouns for case and others do not, thus causing differences in whether or not sentence elements are more tightly or more loosely ordered.  Further, although many languages permit some elements to be null, not all languages permit the same null elements.  Finally, connections among sentence constituents are not always adjacent.  Elements of sentences relate to each other across phrasal and clause boundaries, and the rules pertaining to these relationships are not the same in every language.

Implications for teaching
         This interesting set of facts about languages should not be understood simply as an isolated and interesting bit of trivia; rather, it should help us to realize that learners of a new language expect to find it ordered in ways similar to their own.  They learn to expect certain kinds of information at the beginning, middle, and end of sentences, so may misunderstand the relationships among the various elements in sentences they are hearing or reading in the new language.  They may also omit certain words or create redundancy.  And they may not adequately process referential ties within and across sentences, or they may produce ambiguous ties when they speak or write.  Differences among the sub-modules of language cause a variety of comprehension and production errors.  Although one might expect that with time such errors will simply disappear, we can not know when or even whether that will occur.  This information should at least help teachers understand the kinds of errors LEP students will make, and the kinds of syntactic level phenomena that might be problematic for them as they attempt to process spoken and printed English.

 Activity:  Several of the following sentences are taken from The Gooficon by Burt and Kiparsky (1972).  With a partner identify the errors in the sentences, using the above information.

         1. The English language use much people.
         2. Since the harvest was good, was rain a lot last year.
         3. Was a riot last night.
         4. I bought in Japan.
         5. I was suggested by Mrs. Sena to forget about this project.
         6.  And physical geography prefer me more than anything else.
         7.   The people that I saw them were calling for help.
         8. The doctors that he saw were worried about himself.

After you have identified these sentence level errors, discuss how you as a content teacher might assist the students who have made them.

Back to Table of Contents

         We have already studied phonology (system of sounds), morphology (pieces of words), and syntax (how words join to form larger units such as phrases and sentences), but have not yet addressed the level of words, the lexicon. The lexicon is simply the list of words occurring in a language.  We have already discussed lexical items to some extent in the morphology section, but there are also some ways in which lexical structure is addressed in the syntax.  For example, verbs impose a certain syntactic representation upon the sentences in which they are found.  Note that in sentence (15) 'promise' requires that the subject of the infinitival clause is the sentence subject 'Mary', while in (16). 'want' requires that it is the sentence object 'Tom'.

 15. Mary promised Tom to take out the garbage.
 16. Mary wanted Tom to take out the garbage.

        Each language also imposes phonological constraints on what words can occur in its lexicon.  The word 'bleb' is not a word in English, but it could be.  On the other hand, we would find it very odd if 'tzotiloatl' (Nahuatl for 'place of the vulture') were an English word, because we have neither /tz/ or /tl/ occurring together as consonant clusters in English.  Thus, the lexicon is the set of words in a language, but semantics is the study of word and sentence meaning.

         The study of language is incomplete without attention to what most teachers and learners consider its most important aspect: meaning.  Although as language teachers we often say such things as "'Run' means 'to move quickly on foot'," we should more precisely say "'Run' is assigned the meaning 'to move quickly on foot.'  According to the Western rationalist position, to which many of us subscribe, words do not have meaning inherently, but they are assigned meaning by speakers.  'Run' means 'to move quickly on foot' because speakers of English who know the word understand it that way.  Yet there is no reason why we as speakers of English should use the word 'run' as opposed to the word 'bleb' (a non-word), or 'correr' (the word in Spanish).  Specific words are simply arbitrary combinations of sounds that are assigned meanings by speakers.  Of course, this is not the only position taken on word meaning, for in some Eastern traditions certain words are believed actually to contain properties within themselves, such that when uttered they produce certain spiritual effects.  We are most familiar with the constraint expressed in the Old Testament on the use of the word 'Yahweh' for 'God'.  In the Hindu tradition the word 'Om' is believed to resonate with the cosmos in a positive manner when spoken .  And the Sufis and some other Islamic thinkers express the view that Arabic root words are somehow imbued with the essence of the quality or object they represent.

         Although this point may seem somewhat exotic, there is developmental evidence that young children in their creative play believe in the magic of words, that very young children do not separate words from their referents conceptually, and that bilinguals have an advantage over monolinguals in this regard since they learn early to assign two different words to the same object.  For example, when shown a picture of the moon and told it is a 'bleb,' and asked if the thing in the picture is still a moon, the very young monolingual English-speaking child will say 'no.'   But because the Spanish/English bilingual child of the same age has already learned that the moon can be called 'luna' and 'moon,' s/he will say that the thing in the picture, despite its being called 'bleb,' is still the moon.

         Another philosophical issue that must be made clear is how the form/meaning distinction is viewed.  This is a very important issue in the teaching of English, foreign languages, or English to speakers of other languages.  In some ways we all see meaning as separable from form - represented in such statements as "The content of your paper is weak, although it is grammatically correct."  Such a view may lead us to believe that one is more important than the other, as demonstrated in the following two opinion statements:  1)  "It doesn't matter how I say something as long as you get my meaning," where the speaker implies that meaning is more important than form, or 2)  "Your ideas are unacceptable unless you can express them using correct grammar," where the speaker implies the opposite.  However, in another way, meaning and form are really inseparable, since meaning is encapsulated in forms, the proper ordering of words in sentences being one example.  Each language imposes a certain principled ordering of elements without which it is impossible to construct meaning given just a string of words.  It is not possible to speak, learn or teach a language in the absence of its form, or grammar.  Thus meaning is assigned not only to words, but to sentences and larger units as well.

         Several kinds of semantic relations exist among words, which may be related in these ways (O'Grady et al, 1993, pp. 212-213):  synonomy (same meaning: big/large), antonymy (opposite: wide/narrow), polysemy (many meanings: bank), and homophony (same sound: bear/bare).  Categorizing words in these ways can be valuable in helping learners extend their vocabularies.  Flashcards and board games help students learn on their own, in pairs, or in small group settings.

 Activity:  Make a list of 5 words in your field that are synonyms, 5 that are antonyms.

         Word meaning is closely related to how concepts are represented in the mind.  An old way of thinking of concepts is that there are necessary and sufficient conditions for an object to be called a certain thing.  For example, the defining conditions of 'bird' are generally thought to be something like [+ animal + small + flies] although we would have to admit that ostriches are birds though they are neither small, nor do they fly.  A more recent view is that concepts are fuzzy in that they lack clear-cut boundaries.  For example, if you try to state the necessary and sufficient conditions for something to be a chair or a game or a bird, you will find some "gray areas."  Certain fossil forms demonstrate a biological link between the ancient reptiles and birds.  Are they then birds?  Are Ring around the Rosy or London Bridge games in the same way that Monopoly and football are games?

         Furthermore, while concepts are fuzzy within a language, they become even more problematic across languages.  For example, the word 'aunt' in English may represent any of five relationships: mother's sister, father's sister, mother's brother's wife, father's brother's wife, or close family female friend of the parents' generation.  However, no such generic word as 'aunt' exists in Vietnamese, for there are different words for the various people standing in familial relationships according to the following criteria: the side of the family (mother's or father's), birth status in the parents' respective families (older or younger sibling of the parent), or relationships by marriage.  Although it may seem strange to have to learn all of these possible relationships, it is essential for the young child to do so, for every person s/he meets is addressed using the label of one of these relationships.  Instead of saying the word 'you' for all people as we do in English, the child must learn to refer to his or her elders as if they were members of the family (younger woman 'co,' older woman 'ba,' etc.

 Activity:  Ask your conversation partner what concepts are represented with only one word in English, but must be represented by several words in his or her language.  Similarly, find out if there are concepts represented by one word in his/her language that must be represented by several words in English.

         Another important aspect of word meaning is metaphor, not simply a literary device used by writers, but a pervasive feature of all languages.  Examples in English include the commodity metaphor for talking about time (we waste, save, spend, and lose time) and the conduit metaphor for 'packaging' mental activities (we get points across, put our thoughts into words, and get through to each other).  As we learn new languages we are often charmed by the metaphors expressed, without recognizing that they abound in our language as well, and may be difficult for learners.   For example, in a particular language, time is not a commodity, so one could not translate "waste time" into that language and make sense!

 Activity:  Make a list of metaphors using the names of body parts in English.  What generalizations can you make about any of these?

Activity:  Discuss metaphorical language with your conversation partner, asking him/her to identify metaphors that seem interesting or unusual about English, or identifying metaphors in his/her language that do not exist in English.

 Semantic relations also exist among sentences as shown in the following:

 paraphrase one statement having a similar meaning as another
           The cat chased the mouse. The mouse was chased by the cat.
 entailment one statement entailing the consequence of the second
           The hunter killed the bear. The bear is dead.
 contradiction one statement saying the opposite of the other
           Charles is a bachelor.   Charles is married.
 presupposition a statement made with the assumption that something is true
           When did you stop beating your wife?
 deixis  elements in a statement implying the location of the speaker
           A bear is going into the tent. A bear is coming into the tent!
 speech acts associations with the utterance of a statement
           'I can't reach the salt' being an indirect way of telling someone to pass it   'I pronounce you man and wife' being an act that makes something happen

        When a person knows a language well, these various relationships within and among sentences are usually quite obvious, but they are not so obvious for language learners.  For example, indirectness can be difficult for learners.  When they do not respond to your "It's really hot in here" by offering to open the window, they may not have understood your unstated request.  Non-native speakers may also appear rude when they say something in a direct manner that native speakers express indirectly, such as "I want you to give me x" rather than "I could certainly use x."

         Finally, there are ways in which language, meaning, and thought are so closely bound that it is difficult to separate one from another.  As you may recall, the Whorf-Sapirian hypothesis led to such claims as "We can't think like the Hopi people because we have an entirely different linguistic structure," although it has been strongly refuted in research on children and adults from different cultural backgrounds.  However, a more subtle set of findings has emerged in recent years.  A careful examination of English and Spanish demonstrates that while a large set of English action verbs encode both action and manner (roll, limp, slide, and stroll), Spanish verbs encode action and path (bajar [go down], entrar [go in], subir [go up]).  Similarly, while English encodes aspect (progressive, perfective) on the verb, Hebrew and Vietnamese encode them by other words, usually adverbs.  Finally, while in English location is encoded as prepositions and adverbs, in Swahili locatives are noun suffixes or verbs, and in Korean locatives are nouns!  Such differences surely have an impact on the language learning process.

 Activity:  Consult your conversation partner about how location is expressed in his or her language.

Back to Table of Contents

         Discourse refers to linguistic phenomena beyond the level of the sentence.  The study of spoken discourse includes a wide range of speech activity, i. e., conversation, classroom interaction, turn-taking routines, teasing behavior, or lectures.  Written text can also be discourse.  Discourse studies focus on  how the integrated whole of language is actually used.  Some recent research has focused on the use of how certain discourse routines orient a listener and convey meaning through making the relationship of ideas clear.  Such studies have relevance to our classrooms, especially at the middle and high school levels.  Specifically, certain discourse structuring cues appear to be used by native speakers to make topics clear to a listener, cues that can be taught to students at intermediate and advanced levels to help them understand and  be more easily understood in the content classroom.  Examples include numbering major ideas: ìThere are four main ideas we are going to cover todayî; introducing a topic by using synonyms: ìToday we are going to learn about syntax, that is, the way words are ordered in sentences of a language"; creating logical connections among topics: "Although we have been talking about first language learning, we will learn that many principles of second language learning are similar."

Multicultural issues
         It is at the discourse level that many misunderstandings occur.  Cultural norms are very much involved at this level. For example, the way a discourse is structured between people of different ages, issues of who can interrupt who, how questions are asked and answered, and even what topics are appropriate for discussion are all discourse issues.  Cultural expectations which may be largely unconscious also play an important role at this level.  For example, in some cultures it is not appropriate for children to ask questions of their elders.  By contrast, in most classrooms the United States, students are routinely asked if they have any questions.  However, there are certain expectations dealing with the types of questions that are appropriate, the way these questions can be asked, and even the response of the teacher and other students.  Many LEP students learn to ask questions according to these unwritten rules of classroom discourse, but some do not.  For example, students who comment 'on topic' are viewed as attentive and interested in the topic, while those who never speak, or who speak 'off-topic,' are viewed negatively, despite the fact that their silence or their off-topic comment might belie a deeper level of processing of the topic.

 Activity:  See if you can come up with three questions that would not be appropriate to ask a teacher in an American classroom, but that might be appropriate to ask in another context.

         A typical school event, the parent-teacher conference, has its own prescribed discourse structure which includes types of appropriate topics and responses.  For example, in one conference a parent was told that her child spoke up in class and asked questions, to which he replied, ìOh, I apologize for my daughter.  It will not happen again.î

 Activity:  What does this brief discourse say about the expectations of the teacher and of the parent?  What does it say about their fundamental beliefs and values of what should happen in the classroom?

         Classroom talking behaviors differ across cultures, and discourse patterns are established in the home very early.  For example, Korean mothers teach their babies routines that incorporate familial relationships, while white middle class American mothers teach the question/response routine, a pattern that becomes formalized in school as the "recitation script" (Tharpe & Gallimore 1987); learners supply the answer the teacher has in mind, and are then evaluated.  Among some cultural groups in the United States, however, the QRE routine is quite unusual, and young children may "break the rules" of this script.  Another way in which children may violate majority culture discourse norms is in turn-taking behavior.  While strict turn-taking routines are observed among some ethnic northern European cultural groups, everyone talks at once in others.  Studies of cultural groups as disparate as Eskimos, native Hawaiians, Chinese Americans, working class Christian whites, and African Americans have revealed that each community's patterns of interaction are deeply rooted in their cultural values and beliefs.  Thus, an understanding of such patterns is crucial to the adequate education of linguistic and cultural minority group children.

 Activity:  Consult sources in the reading list that pertain to different cultural groups, and find out discourse patterns of the cultural group you have chosen to learn about.  (For example, see readings by Ladson-Billings, Au, Crago, and Heath.)

Language socialization
         The main way in which we socialize children is through talk.  The ways in which various cultural groups talk with and to their children vary greatly across and also within cultural groups, as we shall see in the brief descriptions given below.  As you read these descriptions, think of ways in which you think you and your brothers and sisters have been socialized through language in your home and community.

         In a recent doctoral dissertation the researcher found that when Korean mothers talk to their young children, they tend to stress the importance of the relationships the child has with other members of the family and community.  They do so by stressing relational terms, which usually occur at the ends of utterances in East Asian languages.  These terms are used to show the relationship between the people speaking together.  For example, the mother may be showing the child a book of pictures in which there is an old woman and a child talking and doing some activity together.  The mother will take both roles, and as she takes the child's part, she will use the appropriate relational term; likewise when she takes the part of the old woman.  In this manner she teaches the child to understand his or her relationship to others and demonstrates in this gentle way that the child will be expected to use those same terms.  Interestingly, the Korean mother is not much interested in teaching the child new vocabulary words, as the middle class white American mother tends to do.  She may even make up words for objects, attributes, etc., which American mothers tend to stress in their talking with children.

         It has been widely reported that the middle class American mother uses a different kind of dialogue with her very young child.  This is characterized by the question/answer scenario, where the mother asks information questions of the child, questions that tend to stress knowledge about the characters in a story, their names and attributes.  For example, she might be looking at a picture book with the child and asking him or her to show a picture of a horse, or to name the animal, tell what color it is, or count the animals in the picture.   Even before the child can speak, the mother role plays this scenario, asking and answering information questions herself.  This pattern has also been observed with some father-child interactions.

         According to Bambi Schieffelin's research, the mother in a rural agricultural village in Botswana rarely talks to her child during the day, because she is working in the fields or selling produce at the market.  Language socialization occurs between the child and older siblings, although these siblings do not necessarily engage the child in conversation at all.  Instead, they tend to teach the child what to say, and ask the child to repeat it.   They might say something like this:  "Say 'I want some sugar in my milk.' "  And the child repeats something like /wasugamiw/, which is considered wonderful and very funny by the older children.  The point here is that children are not always considered conversation partners by adults.  Rather, adults tell the children what to do or say, and they are expected to comply.  In this manner the child is socialized into his or her community.

         According to Shirley Brice Heath, the African American child in a working class community of the Piedmont region is socialized to talk in a different way.  This child may be asked questions, but the questions are often asked to test the child's ability to come up with a clever or insightful response, rather than to respond to questions about known information.  Analogies are valued by the community.  "What's that sound (a mocking bird singing on a wire) like?"  A response like, "Somebody don't know who he is" would be an extremely well-valued response, for the child demonstrates that he understands something about life at a non-superficial level.

         The child also hears a great deal of "ritual" language, namely language that is uttered to create an effect, rather than to communicate meaning directly.  For example, in church, the preacher often asks for responses, "Let the people say 'Amen!'  And they respond, often without being asked.  The rhythm and unity created in the statement-response pattern has a particular effect, again without the communicative value of the words themselves being the primary meaning to be gained from the activity.  As another example, older children and adults might engage in a kind of teasing behavior that calls for one person to insult another.  The other person then comes back with a better insult, the point being to make the other person either to be at a loss for words, or to "lose his cool."  Years ago, this used to be called 'Playing the Dozens,' although every generation has its own name for the practice (i. e., snapping).  The particular African American children in the community studied by Heath are not socialized into the questions/answer routine so prominent in white middle class families, and are therefore not as well prepared to engage in this routine in school: the dominant question/response/feedback scheme, also known as the recitation script so widely used in American schools and reported by Tharpe and Gallimore (1988).

         In many communities the rule that children should be seen and not heard is prevalent, and the children do not initiate questions in the presence of parents and older relatives.  In my own research a Puerto Rican child, Jesus, came from a family with a strict father who was a guard at a local correctional facility.  When told by the teacher that Jesus did not often speak up in class, he said in a conference, ìI have trained him only to speak when spoken to."  The teacher reported being in a quandary about confronting this attitude as cultural or personal, and chose to approach it with caution.  Here is an instance where personal, rather than cultural values, determine the way in which the child will be socialized through language.  For those of you that come from homes where Puerto Rican Spanish is spoken, you might want to compare you own experience with that of Jesus.

         Among some Native American tribes children are neither encouraged nor discouraged from speaking up, but are usually expected to observe and listen to others, and to gradually begin to participate in the community as they become able.  However, Martha Crago and others report that language socialization is changing among the Inuktikut of Northern Canada toward the white middle class norm, as children encounter non-native teachers in their early schooling.

         In Hawaii, native Hawaiian children are socialized to be actively involved in talk along with others in their communities.  "Talk story" is the kind of activity in which everyone is involved in telling a story; everyone talks at once.  If such behavior is allowed in school, and channeled effectively, it can be a good way to build on the kinds of strengths as collaborative learners these children bring to school.   Similarly, Jewish American family talk is often like this, with everyone talking at the same time, interrupting each other, and carrying on several conversations simultaneously.  Some of Woody Allen's movies illustrate this brilliantly.  This kind of talking pattern is also typical among certain Spanish-speaking communities in the Carribean.  Everyone talks at once, and a great deal of emotion is expressed openly.  Many Italian American families similarly incorporate this pattern.   By contrast, a typical Scandanavian community in Minnesota might be more closed, turn-taking routines might be observed more rigidly, and people might express less outward emotion.  A stereotypical version of this pattern was depicted in the movie Fargo.

 Activities: Discuss the following questions with the members of your small group.
         1.  Think about  the typical socialization pattern that you have observed as adults interact with children in your own family.  Does the pattern seem similar to any of those mentioned here?
         2.  Now think about the typical interaction pattern in your classroom.  Does it appear to conform with the same pattern you identified above?  If not, where do you think it came from?
         3.  Which of the above patterns seem most typify instruction in your school as a whole? Are there differences from one of our participating schools to another?  What basic American or community cultural values do you believe these patterns reflect?
         4.  Can you now think of ways in which children of different ethnic or racial groups might have certain expectations regarding the role and value of different patterns of talk than those held by the mainstream school culture?

Back to Table of Contents

First language effects
         In order to account for how people learn second languages, two basic explanations (with various permutations) have been offered.  The first is that first language knowledge is both facilitative of and detrimental to second language learning.  Learners experience interference at the levels of phonology, morphology, and syntax.  Further, if the first and target languages are related, learners may try to use lexical items from the former when speaking the latter.  The interference, or transfer, theory has its roots in behaviorism, thus is accompanied by certain basic assumptions.  First, it assumes that since first language is learned by imitation and practice, second languages should be learned the same way.  It assumes that language learning is forming new habits of speaking, and it assumes that errors in the use of the target language (TL) can be traced back to the first language (L1).

         In the early days of behaviorally-oriented teaching methods, such as the audio lingual method, the teaching of a TL to speakers sharing an L1 (i. e., a group of Japanese speakers learning English or a group of Americans learning Spanish) was based on a contrastive analysis of the two languages: where they were the same, language learning should be facilitated, where different, impeded (predictions were actually much more sophisticated, but there is not time to go into them here.)  Contrastive analyses were usually conducted on surface structure phenomena of languages, as opposed to deeper or more abstract structural phenomena.  More recent work in L1 transfer seeks for cross-linguistic transfer in the latter.  Although many teachers in the language field no longer hold to a strong version of the transfer theory, much language teaching is still based in many of the practices of that time: attention to speaking and oral drill, pattern practice, heavy emphasis on grammatical accuracy, and greater attention to form than to meaning.

Target Language Complexity
         With the advent of Chomsky's 'revolution' in thinking about language as innate and learning as internally driven by more creative processes, a newer theory about language learning emerged in the 1960s and '70s.  As researchers discovered more and more parallels between the errors made by L2 speakers and young children learning their first language, they suggested that more universal psycholinguistic processes influenced second language learning.  Two of these are  simplification and overgeneralization.  Structures produced by early language learners are more brief and structurally simple than those produced by people knowing more of the language.  Structures that are inherently difficult, such as 'do' support, present difficulties to everyone learning English, no matter what their first language.  Certain kinds of linguistic structures seem to be learned in particular orders.  For example, English negation seems to be acquired in the following order:  No, (I) want that (sentence external); She no like it (sentence internal but without auxiliary); We doesn't see anything (auxiliary and contraction present, but overgeneralized form of 'do'); She doesn't have a partner (target, standard form).  As teachers interact with LEP students about their classwork they may be able to recognize their stages of TL development, and should be prepared to adjust their own talk accordingly.  Further reading about early language development is found in the DOE materials, but the following is a brief overview of four stages of second language development that can be used as a heuristic for to plan instruction and adapt content materials.

Back to Table of Contents

Curricular adaptations
         Following is a stage-centered curriculum adjunct that is based on previous reading that posited a four-stage process of L2 learning.  This profile might be used to guide the development of ESL curriculum for K-12 students participating in mainstream content settings.  One of its main flaws is that it assumes simultaneous development of the four skills, and that higher order thinking skills are developed as language abilities grow.  It also assumes that learners will observe a "silent period" before they will want or be expected to speak.  The extent to which stage-centered curriculum planning can be generalized to all situations is also limited by the lack of descriptions at later stages of language development, and it may also not be appropriate for older students with limited literacy skills; it assumes that literacy and conceptual growth occurs simultaneously with language acquisition.

         Nonetheless, once LEP students' levels of English proficiency have been identified, these descriptors may be helpful as a guide to inform the teacher of ways to adapt instruction.  After  this section you will read about the SOLOM, a rating of oral proficiency that includes five L2 stages and that rates speakers on a number of descriptors.   A combination of the SOLOM and this curriculum adjunct should assist you, both to place students in appropriate instructional situations, and to adapt materials according to students' respective stages of  development.

      Stage 1         Preproduction (minimal comprehension, no verbal production)

        Listen, point, move, mime, draw, select, choose, act out, circle

       Stage 2         Early production (limited comprehension, one/two word responses)

        Name, label, group, respond, discriminate, list, categorize, tell/say, answer, count

      Stage 3         Speech emergence (increased comprehension, simple sentences, some basic errors in speech, reading largely limited to what has been learned by ear, writing limited to brief responses, little control)

         Recall, retell, define, explain, compare, summarize, describe, role-play, restate, contrast

      Stage 4         Intermediate fluency (very good comprehension (?), more complex sentences, fewer errors in speech, increased reading comprehension, developing cultural knowledge, errors in written language)

         Analyze, create, defend, debate, predict, evaluate, justify, support, examine, hypothesize

 Activity: Using the list of content classroom-type activities on the next page, decide which would be most appropriate to be used with students at the various levels described above.

 Activity: Select instructional material representing one typical hour-long lesson from your content area and adapt it for some LEP students at stage 1, others at stage 3.

Child first language acquisition
         Weaver, G., Maultsby, S., & Platt, E. J.  (1993). Acquisition of language functions.  Unpublished manuscript, Madison, FL.

         In their first three years of life, young children make remarkable progress in their ability to express themselves using language.  After Noam Chomskyís claims regarding the presence of a ëlanguage acquisition deviceí became known in the early sixties, researchers at Harvard and elsewhere began systematic studies of early child language acquisition within the framework of these claims.  Much linguistics research pertaining to the teaching of second languages is still devoted to exploring the dimensions of developing grammars.  However, in such fields as speech pathology, developmental psychology, ethnography, and human communications, other goals for studying child language development have emerged.  One of these is to trace language development in terms of the ways children express a range of wants and needs.

         Below are three important macro-functions of language in the early years:  Expressing wants and needs relating to objects, actions, and other people.  These data were reported by three North Florida teachers, one of whom recorded speech from her one year-old nephew, another from her two year-old daughter, and the third from her three year-old son.  Notice how the language used to express the various functions changes across the samples: verbal to non-verbal, paratactic to syntactic, one-word to phrasal utterances.  Notice also how much the listener must rely on context to understand the younger children, but needs fewer contextual clues from the three year-old.

Relating to objects
When the child wants something to eat
        1 year  Uhm book.   Uhm cookie. (pointing and/or asking for wanted item)
        2 years  Want that.
        3 years  Give me that. Iím thirsty.   Iím hungry.

When child doesnít want (to do) something
        1 year  No
        2 years  Donít want it.  Donít like it. Mommy do it (shaking head no; facial expression and tone exhibit disapproval of what she doesnít like)
        3 years  I donít want that. No.

When the child wants more
        1 year  Uhm drink. Uhm cookie
        2 years  More juice
        3 years  More juice Iím still thirsty.

Reporting thirst
        1 year  Wahwah
        2 years  Water
        3 year  Iím thirsty.

Wanting shoes off
        1 year  (kicks off shoes)
        2 years  Shoes off.  Socks off.
        3 years  Take these off.

Requesting an object at a distance
        1 year  (takes someone to wanted object, goes himself, or identifies if he knows the word)
        2 years  See um? Want that?
        3 years  I want that.

Wanting the light on or off
        1 year  /ayt/ (sometimes plays with switch)
        2 years  Light off? (off=on)
        3 years  Turn light on.

Relating to actions
Wanting to be moved
        1 year  Up. (reaching hands up) Down.
        2 years  Get down
        3 years  I want down.

Wanting to be rocked
        1 year  Rahrah.
        2 years  Rock rock (showing one finger) more time.
        3 years  Rock me. I wanta rock.

Indicating that s/he is finished eating
        1 year  (drops food on the floor)
        2 years  Through
        3 years  Iím through.

Wanting a ride
        1 year  Tractor  (to label and to ride)
        2 years  Ride truck.
        3 years  I wanna ride the tractor.

Wanting to listen to music
        1 year  (dances or turns radio on and off)
        2 years  Tape on. Mommy song?
        3 years  I donít want to do that.

Wanting to swing or some other activity
        1 year  (remains with activity so as to show he wants to continue it)
        2 years  Swing more? Jump more?
        3 years  I want to swing.

Wanting to go out
        1 year  Out.  (goes to the door and turns the knob)
        2 years  Outside deck? Outside?
        3 years  I wanna go outside.

Wanting to go in a new direction
        1 year  (moves in direction he wants without asking)
        2 years  This way?
        3 years  Uh uh. This way.

Wanting to go to the bathroom
        1 year  (no words yet; probably just does it)
        2 years  Tee tee potty.
        3 years  I need to go to potty.

Relating to people
Wanting help
        1 year  (takes object to someone)
        2 years  Mommy do it?
        3 years  Help me, mama.

Expressing love
        1 year  Ow we
        2 years  Love you
        3 years  I love you, mommy. I love you daddy too.

Wanting to be carried
        1 year  Mommy.  (puts up hands to be held)
        2 years  Mommy hold you?  Kiss you? Hug you? (you=me)
        3 years  Tote me.

Wanting to be read to
        1 year  Read me.  (brings book he wants read to him)
        2 years  Book. Barney. Thomas book. (goes and gets the book herself)
        3 years  Will you read this?

 Looking for someone
        1 year  Calls Papa, Nani, Mommy, Daddy, or Lee (Ayle) when requesting attention or as   if to ask where they are)
        2 years  See Papa? See Granny? See Daddy? See Mommy?
        3 years  Whereís daddy?

Reporting injury
        1 year  (cries)
        2 years  Hurt leg. Hurt knee. Fall down.  (expressing sadness with tone of voice and facial   expression)
        3 years  My leg hurts.

Wanting to know who someone is
        1 year  (does not question who someone is if he has not yet seen them before)
        2 years  Whosat?
        3 years  Whoís that mama?

 Activity:  What do you notice about the development of phonology, morphology, syntax, vocabulary, and sentence length in the data presented above?

 Activity: Talk to your conversation partner or others about their early stages of English language learning, and compare their stages of development with those described here.  If you have been in an environment where another language is spoken, discuss experiences you have had with a limited ability to use the language.

 Activity:  Now examine the acquisition of language functions by a 20 month old child, ìCorey.î  How does he fit into the framework suggested above for the other children?  What generalization are you led to make about the stages of child FLA?

Relating to wants and needs
Wanting something to eat
        whines, cries, reaches in the direction of food
Not wanting (to do) something
        sits down on the floor and kicks his feet
Wanting more
        reaches towards what he wants, sometimes whining or making an ìuh uhî noise

Relating to actions
Wanting to be moved
        wriggling around in the high chair when he wants down, tries to climb out
Finished eating
        begins playing with food
Wanting to go out
        looks out window and bangs on it

Relating to people
Wanting help
Whines in grustration and looks toward adult
Expressing love
Gives hugs and kisses on demand, runs to adult and wraps arms around legs
Reporting injury
Cries, fights back if hurt by another child

Back to Table of Contents

Phonics, Phonology, and Orthography
Data:  The two following lists of words illustrate the problem with "sounding out words" when young children are learning to read English.
ï  How many sounds does the 'ough' spelling produce?
          through, though, bough, rough, cough
ï  How many spellings are there for the /o/ sound?
          doe, throw, fold, home, boat, so, though
Question:  Have your students had difficulties with these sound/spelling problems?  How do you overcome these difficulties?

Activity:  Discover possible alternative sources of Cindy's reading problems.

Facts:
ï  Social behavior:  short attention span, moody, uncooperative, and occasionally disruptive behavior, highly dependent on teacher approval and support;
ï  Cindy's teacher said Cindy could not learn to read because of her attitude.

Problem:  When reading together one day Cindy and I came upon the word 'school' in her workbook, a word she could not read.  When I showed her a picture of a school and asked her what it was, she said "a stool."

Questions:
ï  What is the sound of 'ch' in the word 'school'?
ï  How do you suppose she might have confused /t/ and /k/?  How are the sounds similar?
ï  How would you have helped Cindy?

Related spelling problems:
ï  What sounds does 'ch' make in words of English?
                  choir               Chevrolet         chocolate
                  Christmas       cheroot             church
                  Christ             Chevron           cheese

Problem:  Can you begin to see possible drawbacks in the teaching of phonics to children who speak other languages, or who speak dialects other than Standard English?

Question:  What problems do you think you would experience in doing an exercise in a Vietnamese phonics workbook?

Vocabulary knowledge
         Laufer, B.  (1997).  The lexical plight in second language reading.  In J. Coady & T. Huckin (Eds).  Second language vocabulary acquisition (pp. 20-34).  Cambridge, UK: Cambridge University Press.

Vocabulary knowledge is the strongest predictor of reading success.

Words you (know you) donít know - below the threshold
         ï  in seeking main idea, poor reader discards words not understood, not knowing whether they are important or not
         ï  poor decoder - may know the word if s/he hears it, but does not have letter/sound association skills to figure it out
         ï  limited English proficient - may know the word in his/her own language, but does not know the English word
         ï  threshold (sight) vocabulary
                  words that are processed automatically,
                  not words vaguely understood
                  not words that can be inferred from context
         ï  focusing on unfamiliar words
                  takes up processing time
                  limits cognitve capacity
                  interferes with the ongoing process of knowledge construction and higher level processing of text that the reader is performing during reading
         ï  large automatic voacabulary frees cognitive resources
                  to make sense of unfamiliar words
                  to capture relationships among sentences and paragraphs in the text
                  to interpret text globally
         ï  number of words that are needed (by adults in academic programs) to minimally pass a reading test - 3000 word families, or 5000 single lexical items; increases of 1000 words only improve scores by 7%.  For middle and high school levels the expectation is not this high, though many LEP students in our classes have far fewer words than this.

Words you think you know
         Deceptively transparent words (mistaken identity) can create serious problems for readers if they are not made aware of them.  Some examples: infallible (something that can not fall); shortcomings (short visits).  Specific types:
         ï  deceptive morphological structure - outline (out of line); nevertheless (never less)
         ï  idioms - translated literally (miss the boat, hit and miss, sit on the fence)
         ï  false friends, false cognates - assuming words having a similar form in the two languages have the same meaning (embarrassed, embarasada (pregnant in Spanish)
         ï  multiple meanings - when familiar meaning is seen as having only one meaning; ëstateí understood as country when it means situation; in math table, scales, bar, etc. assigned their non-technical meanings
         ï  synforms (similar lexical forms) - cute/acute, available/valuable, conceal/cancel, reduce/deduce/induce.  Thinking they understand these words, learners assign incorrect meanings to them, thus creating distortion of the immediate context.  Building on this local distortion, they may guess meaning of later unfamiliar words in terms of the schema being constructed around this understanding, and subsequently create higher level distortions of the whole text (les torreaux, les trops).

Words you canít guess
         Guessing strategies may be unsuccessful when the number of known words is limited.  Factors that interfere with guessing include
         ï  non-existent context clues - can not be certain of contextual redundancy
         ï  unusable context clues - when the supportive context also includes unknown words; reading strategies such as using context clues depend on a high level of text understanding to begin with (95-98%!)
         ï  misleading clues - In attempting to understand the following text in Italian,  E dappertuto si bevava, si cantava, si ballava, si rissava . . . (And everywhere (people were) drinking, singing, dancing, brawling . . . ) learners mistakenly assumed ësi rissavaí to be ëenjoying themselves,í since it fit in with the rest of the text.
         ï  partial clues - Typhoon Vera killed or injured 28 people and crippled the seaport of Kellung.  If ëcrippledí were understood as ëdamagedí or ëdestroyed,í the meaning would be partially right, but not precisely.  In math it is almost always necessary to construct precise meanings.
         ï  suppressed clues - drawing on background knowledge is one of the successful strategies learners use to help understand text.  But the background knowledge brought to the text may be so different that learners may attempt to interpret words in terms of their known schemas, thus guess incorrectly (Plattís experience with cows in France)