LEON A. JAKOBOVITS (Ph. D., McGill University) is currently an Associate Professor of Psychology and Communications at the Center for Comparative Psycholinguistics at the University of Illinois.  He has taught at Laval University and is the Co-Director of the Centre for Comparative Psycholinguistics at Illinois.  He is a member of the American Psychological Association, the International Council of Psychologists, and the Linguistic Society of America.  He has published articles in the International Journal of Psychology and the Journal of Special Education.  He is also the co-editor of Reading in the Psychology of Language, published by Prentice-Hall in 1967.

* Based on a report prepared by Leon A. Jakobovits for the Liberal Arts School Committee on the Foreign Language Requirement, University of Illinois, Urbana, 29 May 1968.

    The purpose of this report is to outline some major issues in foreign language teaching today, issues that bear upon the current, profession-wide examination of FL requirements.  This report does not attempt to give a general review of the literature on FL learning and teaching; instead, it relies on previous extensive reviews (see References), taking them s the staring point, and it attempts to outline those principal conclusions which seemed to have the most empirical justification.  The report is organized into six segments:

    1.  Teaching Methods in FL Instruction.
    2.  The Case for FL Aptitude.
    3.  The Attainment of FL Proficiency.
    4.  The Effects of Motivation and Interest in FL Learning
    5.  The Goals and Benefits of FL Study.
    6.  Recommended Changes in FL Requirements.
 

    Two theoretical approaches take up most of the attention of language teachers today.  One, the habit-skill approach, views language behavior as a chain of habit units, the main problem in teaching being defined as making these habit chains automatic.  The other, the rule-governed grammar approach, views language competence as the ability to generate nobel utterances on the basis of finite set of rules, the main problem in teaching being the impartation of an adequate knowledge of these rules.  The habit-skill approach emphasizes oral practice techniques with a minimum of explanation of grammatical rules (such as repetitive sentence pattern drills).  The generative approach discounts the usefulness of traditional concepts in learning theory such as practice, conditioning, reinforcement, and concentrates instead on arranging the linguistic input to the learner in a way that would maximally facilitate his acquisition of structural patterns and rules.  It should be made clear that emphasis on rules and structure of the generative approach is not to be confused with the traditional (now archaic, except perhaps for the teaching of Latin) approach of the so-called grammar-translation method, which attempted to impart to the student a knowledge of the formal grammatical distinctions between his native language and the target language.  In the generative approach, the structure and rules of the target language are imparted inductively with out the use of a formal meta-language and without translation exercises.  Both the generative approach and the habit-skill approach may make use of modern linguistic theory on the nature and structure of language relying on generative transformational theory and contrastive analysis.  Both tend give greater emphasis to speaking and understanding, but the extent to which reading and writing skills are also taught varies greatly within either approach.
 The debate on the relative merits of the two approaches, which occupies a significant share of the tension of FL teachers and applied linguists in their deliberations at conferences and literature, is not nearly of academic interest.  The polemic character of the debate is kept alive by two disturbing aspects FL teaching today: one is the wide-spread dissatisfaction expressed by both students and teachers with the repetitive pattern drills required by the habit-skill approach, which makes the former feel like idiots and the latter more like drill masters than teachers.  The other reason has to do with distinct malaise that came to permeate the ranks of FL teachers upon the realization that the level of proficiency attained by a large, if not major, proportion of public school and college FL students is disappointingly poor.  Given this situation, it is necessary to examine in some detail the nature of the arguments in this debate.
 Some standard arguments were recently outlined in a critique by Bernard Spolsky (1996) of the Linguistic department at New Mexico.  He summarizes the major assumptions upon which the habit-skill method is based:

    1.  FL learning is a mechanical process of habit formation.
    2.  Habits are strengthened by reinforcement.
    3.  Language is behavior made up of habit sequences at the phonemic, morphological, lexical and
         syntactic levels.
    4.  Repetition, practice, and reinforcement of units and their concatenation are effective ways of
         developing language performance.

These principles are said to be derived Harvard psychologist Skinner’s behavior theory.  Skinner’s (1957) major attempt to apply the principles of behavior originally derived from work with rats and pigeons to the language behavior of humans has been persuasively demolished by Chomsky (1959) in his review of Skinner’s book Verbal Behavior.  In this review, Chomsky shows that the operational concepts in Skinner’s theory of verbal behavior are gratuitous extensions from his work on animal behavior and lose their explanatory power completely when applied in the context of human verbal behavior.  For example the concept of “response strength”- his basic measure of learning- is defined in the work with rats as number of bar presses in a box rigged to deliver pellets of food upon depression of a level by the animal.  In the pigeon, response strength is equated with number of pecks.  But what is it in language behavior?  Frequency of emission of words, rate of speech units, intensity of vocal response, and the like, have been proposed but all these are clearly inadequate (think of the absurdity of the conclusion that a language known less well when it is whispered-low response strength- than when it is shouted!).  Another example concerns the concept of “control,” which is said to be the end result of the learning process, viz., when a particular “response” is “brought under the control” of a particular “stimulus” through the consequences of the response (i.e. reinforcement).  These processes which have operational meaning in the animal laboratory (e.g. stimulus = a light, response = bar press, control = presence of bar press when light is on, reinforcement = pellet of rat food), lose their specificity and become completely obscure in language behavior and can be said to apply to this new area only by analogy and metaphor, hence devoid of scientific explanatory value (think of what might be the “stimulus” of an utterance, in what sense an utterance is a "response," and by what means can one say that one "controls" the other through what reward).  Yet it is precisely these specific tenuous extensions upon which the habit-skill method of language teaching is said to be based by its proponents (see Rivers, 1964).
    With respect to the role of repetition and practice, recent investigations of natural language acquisition in children clearly show that overt practice on specific grammatical examples carries little or no weight in "implanting responses" and that "imitation" of novel grammatical forms occurs infrequently (for a detailed discussion of this argument see Jakobovits, 1968).  Thus these concepts (repetitive practice, implanting of responses, imitation), cherished and carried to the ultimate in repetitive pattern drills, have only the most tenuous scientific justification.  The habit-skill approach to FL teaching curiously rests its justification on a sequentially controlled model of language, despite the fact that such an approach has been clearly refuted in both psychological (see Lashley, 1951) and linguistic (see Chomsky, 1957) literature.
    It is important to consider the implications of the above critique for a new development in FL teaching, namely programmed instruction.  Several instructional programs are now commercially available in the form of language texts for French, Spanish, and German.  The invention of programmed instruction is attributed to Skinner (the "father of teaching machines"), and the basic scientific justification upon which this new technique rests is similarly related to the principles of behavior he developed in his work with animals.  All the difficulties discussed above apply here with equal force, but there is even an added complication: programmed instruction requires the isolation of "units" which compose the competence to be taught or the knowledge to be acquired.  It has been clearly shown (Chomsky, 1965; Lenneberg 1967, and many others) that the significant knowledge a user of a language has to acquire does not constitute "units" but patterns and relations.  This holds true for all levels of linguistic analysis: phonological, semantic, syntactic, and morphological.  For example, recent research has shown that a phoneme is not a physically distinct acoustic unit: it cannot be taught as a "unit," only as a class of variable phones which have certain relations to each other as well as to other classes of phones.  Similarly, one cannot acquire "true" language competence by learning specific grammatical patterns as "units" since the number of sentence patterns understood by a native speaker is infinitely variable-one cannot seriously hope to teach true language competence by mechanical mastery over a limited number of sentence patterns.
    However weak the theoretical position of the habit-skill approach may be, one can nevertheless inquire as to its comparative effectiveness to that of the generative rule or grammar approach.  Numerous studies in the last twenty-five years or so have attempted to resolve the issue of effectiveness of various methods of FL teaching, none of which has been adequate enough to permit any definitive conclusions.  The reason for this disappointing state of affairs is that it is practically unfeasible to vary one element of instruction experimentally without at the same time modifying the effects of other elements in an unknown manner.  Hence, as in all complex educational problems, it is perhaps unwise to expect a scientific assessment of language teaching methods.  In a recently completed extensive project carried out at the University of Colorado by Scherer and Wertheimer (1964), an ambitious attempt was made to compare the effectiveness of two different instructional programs for teaching German in college: one approach emphasized audiolingual skills while the other method emphasized reading and writing.  At the end of one year of instruction, the audiolingual group was found to be "far superior" in speaking and listening skills, while the more traditionally trained group was "significantly better" in reading and writing.  During the second year, the two groups were merged and given common course instruction.  At the end of the year, the first group was better speaking while the second group was better at writing, neither of them differing in listening and reading.  The authors concluded that "the two methods, while yielding occasionally strong and persisting differences in carious aspects of proficiency in German, result in comparable overall efficiency."  The results of this and several other studies of the same scope reviewed by Carroll (1966, 1965) lead him to “the rather commonplace conclusion that by and large, students learn (if anything) precisely what they are taught” (1965, p. 22).
 Many other experiments of lesser scope have been carried out in attempts to evaluate relative effectiveness of various methods of teaching phonology, grammar, and vocabulary, but it is not within the purview of this report to give an evaluation of these attempts.  The interested person is referred to the review by Carroll (1966).  Attention is brought to recent efforts to develop self-instructional programs, especially to the work of Valdman (1964), who has proposed the idea of giving college students FL credit proportionate to their level of achievement rather than to the amount of time they spend taking courses.  Carroll is very optimistic about the possibilities of such a program: “It is evident that self-instructional programs in foreign languages are not only perfectly feasible but also highly effective-more effective, in general, than conventional teacher-taught courses.  When used by sufficiently well-motivated students, they can produce high levels of attainment in all four skills of language learning-speaking and writing as well as listening as reading” (Carroll, 1966, pp. 27-28).  It should be made clear that the effectiveness of programed self-instruction has not been adequately assessed for the general college population.  It is known that individuals with different aptitudes and rates of learning require adjustment in learning programs and the technical problems involved in adjusting FL self-instruction programs to the individual learner have not been worked out

    In order to be able to evaluate the manner in which learner characteristics, such as aptitude and motivation, interact with methods of instruction, it is necessary to consider the definitions for certain terms that will be used in this discussion.  These definitions are based on Carroll (1963a).

     Learning: a task to be acquired; performing an act which previously could not be accomplished;
                     understanding a concept previously not understood.
     Transfer: when something learned in situation A also manifests itself in situation B because of the
                     inferred commonality between the two situations.  (The elements in common are often
                     not specified or even understood.)
     Learning time: amount of time spent on the act of learning.  (Not to be confused with elapsed
                             time which includes such activities as sitting at the desk dreaming, “wasting time:
                             looking for a book or pencil, etc.)
     Aptitude: learning time under best teaching conditions; the shorter the learning time the higher the
                    inferred aptitude.  (Note that for a difficult task combined with a low aptitude, learning
                    time may be indefinitely long.)  Aptitude is specific to tasks and depends on possession
                    of certain characteristics by the learner.  These characteristics may be either genetic
                    (innate) or they may be dependent upon prior learning or exposure to certain
                    situations.
     Ability to understand instructions: this is conceived of as dependent upon two factors: general
                                                            intelligence and verbal ability.  The first enters into the ability
                                                            of the learner to infer the concepts and relationships needed
                                                            for the task-especially when these are not carefully spelled
                                                            out (which, one might add, is the usual situation in any
                                                            complex learning task).  The second comes into play in the
                                                            understanding of the language used in the instructions.
     Quality of instruction: the extent to which it is made clear to the learner what it is he is
                                         supposed to be learning.  (Note that this refers to highly specific
                                         elements within the overall learning task.  Thus, telling the learner that he
                                         is supposed to acquire “ a reading knowledge of this FL” says absolutely
                                         nothing about what he is supposed to be learning: does he begin by
                                         learning the “writing system,” “vocabulary items” (which ones?),
                                         sentence patterns, phonology (how?), etc.  Even in sub-tasks such as
                                         listening comprehension, the learner is confronted with the problem of
                                         just what he is supposed to be paying attention to: phonemic contrasts,
                                         segmentation, contour, grammatical relations, etc.-  It is evident that
                                         “quality of instruction” deals with such highly complex (and unsolved)
                                         problems as the identification of relevant contrasts, their sequencing, the
                                         amount of exposure needed at each level, and so on.  It should also be
                                         noted that, as defined here, quality of instruction does not have an
                                         absolute criterion but relates to the learner's point of view, viz., whether
                                         or not the task has been made clear to him.  His aptitude, previous
                                         knowledge, and ability to understand instructions will influence the
                                         specific requirements for making it clear what he is supposed to be
                                         learning.  It follows that "a standard" method of instruction which does
                                         not vary with the learner's characteristics could not be of high quality.
    Perseverance: the time learner is willing to spend in learning to a specified criterion.  (This
                           definition may not be quite adequate since a learner may "be willing" but "is
                           unable" due to distraction, frustration, etc.  However, to resolve this issue one
                           would have to go into a discussion of the type, Is he really "willing" when he
                           "allows himself" to be distracted?  etc.-a type of discussion which would take us
                            too far afield from the central purpose of this report.)
    Opportunity to learn: the learning time allowed by the method of instruction and the
                                      environmental conditions.  (Applies to sub-steps as well ass to the overall
                                      task.)

    We have now considered the definitions of all the concepts, stated or implied, required in the following formula given by Carroll:

The numerator in this equation will be equal to whichever of the following three terms is smallest: opportunity to learn, perseverance, instruction and ability to understand instructions.
    The problem to which this report is addressed concerns, of course, a review of the factors that enter into an evaluation of the degree of learning of FLs in the school situation, in particular the college or university.  The discussion in the previous section on methods of instruction is particularly relevant to two concepts implied in the equation: quality of instruction and opportunity to learn.  The next section on the measurement of proficiency deals with a specification of the criterion to be achieved, which has not been dealt with so far.  The fourth section on motivation will deal with perseverance as defined here.  The rest of the present section will deal with the remaining two concepts implied in the equation: the ability to understand instructions and aptitude.
    It is known that the acquisition of any learning task is influenced by what has come to be called a "general intelligence factor."  As defined here, ability to understand instructions is a joint consequence of this general intelligence factor and verbal ability.  The challenge involved in developing programmed instruction can be viewed as an attempt to reduce the limiting effect of the general intelligence factor by sequencing the learning task in sufficiently small and clear steps so that the learner doesn't have to depend on his ability to guess just what he is supposed to be learning.  It is this feature rather than its purported relation to Skinner's behavior theory that makes this new development such an exciting prospect.  But the "state of the art" in programmed instruction, particularly in FL teaching, is at such a crude level that the learner's general intelligence remains an important limiting factor in degree of learning.  It is not surprising, therefore, that grades obtained in FL courses correlate with grades obtained in other school subjects.  In a survey conducted in 1965 at the University of Illinois (Flaugher, 1967) involving the grades in nine fields for all entering freshmen, the interrelations at the end of the first semester of work ranged from .18 to .66 with mean of .38.  Intercorrelations involving FL grades ranged from .34 (with Speech) to .51 (with Natural Science) with a mean of .42.  This last figure is of the same order as that found by other surveys (see Pimsleur, Sunland, and McIntyre, 1964).  These figures indicate that, while general intelligence is a contributing factor to success in courses, it accounts for only a modest proportion of the variance in success (a correlation of .42 accounts for 18% of the variance, leaving 82% of the variance to be accounted for by other factors).  Indeed, as is well known, many students who obtain high grades generally do very badly in their FL course, and vice versa.  This has led many to postulate "a special talent" for FL study.  During World War II the U.S. government became involved in FL training programs designed to impart a "practical speaking knowledge" of many FLs to personnel assigned to overseas duties of a diplomatic, military, and paramilitary nature.  These programs, known as "intensive FL training programs," apart from their costliness, were associated with a sense of wartime urgency which lead to efforts of selection of trainees to "weed out" from the beginning those persons who had insufficient talent to complete the intensive program successfully.  These initial attempts at prediction of success in FL study found that the best predictor of success in an intensive program was the student's performance in an intensive trial course that usually lasted three to four weeks (Carroll, 1965).  On the basis of this experience, tables of expectations of success were developed, one of which is presented here as Table 1 (p. 454).
    The table appears in Cleveland, Mangone, and Adams (1960, pp. 250-251) and estimates the time requirements (in months) for FL achievement in intensive programs for individuals with "high" and "average" aptitude.  Three levels of proficiency are defined as follows:

    Level I: "Sufficient proficiency in speaking a foreign language to satisfy routine travel
                  requirements."
    Level II: "Basic familiarity with the structure of a language with sufficient proficiency in speaking
                   to conduct routine business within a particular field.  Sufficient familiarity with the
                   writing systems to read simple material with the aid of a dictionary."
    Level III: "Fluency and accuracy in speaking with sufficient vocabulary to meet any ordinary
                     requirements which do not involve the speaker in a technical subject outside his own
                     specialty.  Ability to read newspapers and documents with limited reference to a
                     dictionary."  The time estimates are further divided into three degrees of intensiveness
                     of the program: "1 Class Hour" refers to one hour of instruction per day supplemented
                     by "2-3 hours of drill and study."  This "low" level of intensiveness is not considered
                     practical as an intensive program to go beyond Level I.  (Note, however, that it is far
                     more than the usual college FL course.)  The target languages to be learned are
                    divided into four groups.  The "easiest" in terms of time requirement (for native
                    Americans) include the most frequently taught FLs in American schools (Spanish,
                    French, German) but also include languages that few college students normally think of
                    taking (Dutch, Danish, Swedish).  The next group of languages in terms of difficulty
                    includes Russion, which is becoming more popular, and Modern Breek, as well as
                    more "exotic" languages such as Persian and Hungarian.  The most difficult languages
                    for Americans are the "Asian" varieties including Chinese, Japanese, Vietnamese, and
                    Thai, but also Arabic.  The difficulties involved in mastering the writing system of
                    these languages (at Level III) add a very appreciable amount to the time requirements.

    The superiority of the "high aptitude" individual over persons with "average aptitude" can be seen to be considerable (people with "low aptitude" are not considered capable of attaining any significant level of achievement under the time requirements of an intensive program).  For the easiest languages, the "high" aptitude person can attain Level III proficiency in one-half to two-thirds the time needed for the "average" aptitude individual.  For the difficult languages, the time is between 20% and 33% faster.  Carroll (1960) estimates that only one-third of the general population in the U.S. has a sufficiently high degree of FL aptitude to complete successfully intensive language training programs designed on the order of 400 hours of study (a four-year high school FL program represents on the order of 600 hours of study).
    As aptitude was defined earlier in this section, it is clear that a low level of aptitude can be compensated for (up to a point) in two ways: increasing time for opportunity to learn and lowering the level of proficiency to be attained.  It can be seen that, assuming adequate motivation to learn (perseverance), and fixing an upper limit for time of study-say four years of school work-one can set the maximum level of proficiency that individuals with a given aptitude could attain.  If we now set the minimum acceptable level of achievement-say in terms of a score X on a proficiency test-as the requirement for "success," there will then be a certain proportion of the population that will not achieve success in FL courses.  Pimsleur, et al. (1964), estimate that up to twenty percent of the student population in high schools and colleges are "beset by a frustrating lack of ability" in FL study.  They refer to these students as "underachievers" in view of the fact that their grades in FL courses are "at least one grade point lower than their average grade in other major subjects."  These investigators set about on an ambitious project to identify the characteristics and specific abilities that constitute FL aptitude.  The administered a battery of tests to high school students in a state school system that they considered typical of the Midwest and the Eastern Seaboard (probably in Ohio) and correlated the scores on these tests with obtained school grades in FL courses.  An evaluation of their effort is given by the following comparison of variable that correlate with FL course grades:
 

As can be seen, their Aptitude Battery (which takes less than two hours to complete) is as good as a predicator as Grade point average (which represents a whole semester of work in several subjects).  When these two variables are combined, the multiple correlation affords more than double the predictive value of an intelligence test, and one-half of the total variance to be accounted for.
    It would be instructive to examine the tests of the Aptitude Battery as it may give an indication of what constitutes a :talent for FL's."  Here is a brief description:

    Interest Test 1: A series of questions designed to index how eager the student is in studying the
                             language he is taking.
    Interest Test 2: A series of questions evaluating the student's belief in the general value of FL
                             study.
    Linguistic Analysis Test: "A fifteen-item test of verbal reasoning in which the students are given
                                             a number of forms in a foreign language and asked to deduce from
                                             them how other things are said in that language."
    Vocabulary Test: A vocabulary richness test as a rough measure of verbal ability.
    Pitch Test: "A test of auditory discrimination in which the student must distinguish Chinese tones."
    Rhymes Test: A test to measure fluency with words.
    Sound-Symbol Test: A "rapid-fire test in which the student hears a nonsense syllable and must
                                    match it with the correct spelling in his booklet."

    Not all these tests are equally related to FL aptitude.  In fact, the results of a matched-group experiment, in which "underachievers" were compared to "normal" students, showed no difference between the groups on the "Linguistic Analysis Test," the "Vocabulary" tests, and the "Rhymes" test.  On the other hand, on three of the tests the underachievers scored significantly lower: these were the Interest Test 1, the Pitch Test, and the Sound-Symbol Test.  The conclusion Pimsleur, et al., reach is unambiguous:

    According to this investigation, there does exist a "talent" for learning foreign languages-that is, a special factor beyond intelligence and industriousness which accounts for how well an individual succeeds in a language course.  OUr evidence indicates this special factor is auditory ability, which may be defined as the ability to receive and process information through the ear.  (1964, p. 135)

A similar conclusion is reached by Carroll (1963), whose extensive work on the Modern Language Aptitude Test (MLAT) was carried out independently of and prior to the Pimsleur, et al., investigation:

    These propositions [which this chapter will attempt to demonstrate] are (a) that facility in learning to speak and understand a foreign language is a fairly specialized talent (or group of talents), relatively independent of those traits ordinarily included under 'intelligence' and (b) that a relatively small fraction of the general population seems to have enough of this talent to be worth subjecting to the rigorous, intensive, expensive training programs in foreign languages...[However,]...the question of whether a student of lower than average aptitude should study foreign languages for purposes of general and liberal education depends upon a number of considerations which do not bear upon the selection of students for intensive foreign language courses of the type described here.  (p.89)

    According to Carroll, FL aptitude consists of at least the following four identifiable abilities as measured by the MLAT: (1) phonetic coding, which is "the ability to 'code' auditory phonetic material in such a way that this material can be recognized, identified, and remembered over something longer than a few seconds."  This ability is then very similar to that identified by Pimsleur, et al. (1964), as discussed above.  The word "coding" in the description takes on added significance when it is realized that "phonetic discrimination" per se is not an important predictor of FL success.  A test of the ability to perceive phonetic distinctions by requiring the listener to distinguish between similar sounds presented as "foreign syllables" was included in earlier versions of the MLAT Battery but was later abandoned: its "validity coefficients were consistently low in comparison to those of other tests, and the conclusion was reached that phonetic discrimination ability is not crucial in foreign language learning.  Most normal people have enough discrimination ability to serve them in learning a foreign language, and in any case, it is more a matter of learning the discrimination over a period of time than any fundamental lack of auditory discrimination which can readily be tested in an aptitude battery" (Carroll, 1965, p. 96).  Thus it appears that the popular notion of "having a good ear for languages" is an ability that doesn't depend so much as one's "ear" but on the brain's capacity to code and store for later recall auditory information of a phonetic type.
    (2) The second major ability measured by the MLAT is grammatical sensitivity, which is "the ability to handle 'grammar'," i.e., the forms of language and their arrangements in natural utterances.  The sub-test which measures this ability ("Words in Sentences Test") requires the individual to recognize the function of words in various contexts using English sentences.  For example, a word or phrase in a second sentence which has the same grammatical function as the underlined element in the first sentence.  Thus, the test does not require formal training in the matalanguage of grammar, although such training may improve this trait.
    (3) The third variable measured by the "Paired Associates Test" of the MLAT is "rote memorization ability for foreign language materials," and "has to do with the capacity to learn a large number of...associations in a short time."  It is well known in the psychological literature on verbal learning that rote learning ability is not related to intelligence to any substantial degree.
    (4) The fourth variable in FL aptitude is "inductive language learning ability," which is the "ability to infer linguistic forms, rules, and patterns from new linguistic context with a minimum of supervision and guidance."  Unfortunately, this ability is not measured by the present commercial version of the MLAT.
    The validity coefficients of the MLAT vary greatly, depending on the subject sample and the population they are drawn from.  For example, its correlation with two groups of students enrolled in the Five-University Summer Program in Middle Eastern Languages was .40 in one group and .58 in another.  The course was an eight-week intensive program in various languages including Arabic, Turkish, Persian, and Modern Hebrew.  In further extensive tests of the MLAT, Carroll (1960) obtained twenty-eight validity coefficients for high school courses; these ranged from a low of .25 to a high of .78 with a median of .55.  In twenty-five coefficients obtained with college courses the range went from a low of .13 to a high of .69 with a median of .44.  These coefficients were superior to those obtained with intelligence tests and there apparently "was no systematic fluctuation of validity dependent on teaching methodology."
    An interesting aspect of the MLAT is the claim Carroll makes that it is equally valid for predicting success no matter what the target language is.  The status of this claim, however, is uncertain in view of the ambiguous data available on this question (see, for example, Table 4.11 in Carroll, 1965).  Nevertheless, the evidence for the positive, significant validity of the MLAT in all languages is strong and there is little evidence so far that particular languages require special abilities.
    Another interesting aspect of the MLAT is its potential use as a diagnostic tool.  At the present time, it is not known to what extent one can improve the various separate abilities in FL aptitude by remedial training.  On the other hand, a type of instruction that takes into account an individual's specific weaknesses as revealed by his scores on the sub-tests of the MLAT would seem to be a helpful strategy.  Even if individual attention is not possible, separate classes based on the MLAT as a placement test would be indicated.

    It is not the purpose of this section to review existing tests of proficiency for FLs or the principles involved in the development of such tests.  Rather, the intent is to examine the nature of the goal of FL achievement, what Edgerton and his associates (1968) have called "liberated expression" in their report to the Northeast Conference on the Teaching of Foreign Languages.  An extended quote from their report will introduce the problem to which this section addresses itself:

Most existing materials for classroom instruction in foreign languages-espcially audiolingual materials-concentrate on teaching the overt "machinery" of the languaag....When the question of meaning arises in the context of foreign-language instruction, "culture" must necessarily be considered since the semantic component, the "meaningful content," of a language cannot be separated from the culture of which it is a vehicle....Blunder after blunder (of the very sort the student's experience with a native or near-native "monitor" would sort out for him) is due not to lack of adequate conditioning in the manipulation of the mechanical aspects of the language he is trying to speak, but to a failure properly to associate the "pieces" of that language with what they denote and connote in the foreign culture itself....Very often it is assumed that the primary aim of study of a foreign language in the context of a general education is to train the student so that he can make practical use of his acquired skill and knowledge.  However, on close inspection this aim seems quite unrealistic.  The great majority of students who study a particular foreign language in the course of their schooling never make very much actual use of it for either professional or casual purposes.  (pp.100-101)

FL requirements in colleges and universities, both entrance and graduation, are usually stated either in terms of number of credit hours taken or in terms of some vague statement such as "a speaking knowledge" or "a reading knowledge" of the language.  In evaluating the success of the FL training program in the American educational system, it is necessary, as the above quote suggests, to consider the attained proficiency on two levels.  The first level concerns the mechanical manipulation of the FL, while the second, "higher" level, relates to the student's attainment of "liberated expression"-the ability to use the FL as a vehicle of communication, what might be called communicative competence.
    With respect to the first level, it is well known that successful completion of "so many" course credit units is a very poor indicator of the degree of proficiency attained.  Even standardized tests of FL achievement are ambiguous in this respect for, as Carroll (1960) points out, they allow an assessment of competence only in terms relative to a comparison group which is considered "typical."  Percentile scores do not indicate the absolute level of achievement of a student.  Carroll gives an example of what he considers "a meaningful scale on which it would be desirable to report proficiency" (1960, p. 72).  This scale was established by the Foreign Service Institute of the U.S. Department of States.  The standards in speaking proficiency are:

    S-0:  no practical knowledge of the language
    S-1:  able to use limited social expressions, numbers, and language for travel requirements
    S-2:  able to satisfy routine social and limited office requirements
    S-3:  sufficient control of structure and adequate vocabulary to handle representation
            requirements and professional discussions in one or more fields
    S-4:  fluency in the foreign language
    S-5:  competence equivalent to English

A similar series of standards has apparently been established for reading and writing.  An example of such a scheme has already been provided in Table 1, previously discussed, which also gives, as will be recalled, the time requirements for attaining the various defined levels of proficiency.  As long as school requirements of FL proficiency fail to specify the levels of achievement in terms similar to these, no meaningful assessment of the success of their programs is possible.  Furthermore, until such as assessment procedure is adopted, the requirements for FL study are likely to continue to appear to the student as arbitrary and irrelevant, a condition that is not likely to foster the kind of interest and motivation that promotes FL study (see next section).
    It is often stated that full native language achievement is an impossible goal for FL training in the school.  While such an expectation is undoubtedly unrealistic under present conditions of instruction, considering the limited time available for FL study for most students in high schools and colleges, it need not imply that because full communicative competence is impossible instruction should be geared to nearly a knowledge of the mechanical manipulation of the language.  The various levels of speaking proficiency as defined by the Foreign Service Institute scale (see above) assume the mastery of various levels of communicative competence before complete mastery of mechanical manipulation is attained.  Fore example, level S-4 (fluency in the FL) is preceded by S-2 (able to satisfy routine social requirements), which is short of the proficiency in mechanical manipulation achieved by many FL students in school courses yet is far ahead of the latter in terms of ability to use the language in a practical situation.  It would appear that the low level of communicative competence achieved by students in FL courses in school is due more to a failure in the method of instruction than to lack of sufficient time devoted to FL study.
    Hayes (1965), in a most perceptive discussion, has outlined the "new directions" that FL teaching ought to take to remedy the "gap" between a knowledge of "structural manipulation" and communicative competence.  It is instructive to review briefly the nature of the knowledge that must be subsumed under communicative competence.

    A.  Paralinguistic factors.  These refer to features of speaking such as pitch, range, tempo,
          clipping, etc., which may make one react to a speaker with the comment "It's not what he
          said, but how he said it..."
    B.  Kinesic factors.  Formal, visual (facial, gestural, etc.) features that carry meaning such as the
         slightly protruded tongue whose significance caries not only cross-culturally and
         cross-linguistically but, also, contextually within a culture.  (Consider the difference in meaning
         of this act in our culture when done by a child concentrating on a task, by a young woman to
         her intimate friend, or by a effeminate male.)
    C.  Sociolinguistic factors.  Linguistic features (phonetic, lexical, grammatical) correlated with
         geographical origin (dialect), socioeconomic background, context of interaction (speech
         style), speech mode (writing vs. speaking), etc.
    D.  Psycholinguistic factors.  Features of the communication situation (including all of the
         above) as they affect the listener.  For example, the person who uses the "High" version of a
         language in a speech community characterized by diglossia (see Ferguson, 1964) where the
         "Low" version in required (as in Arabic, Greek, or Swiss German) is an object of ridicule,
         while the reverse blunder in a school situation may land the teacher in jail (in some
         Arabic-speaking countries).  Or, to give a more familiar example, the use of informal
         phonological style by a non-native speaker evokes an unfavorable reaction (e.g. "I'm gonna
         go home," with a less than native accent on "gonna").

    When these aspects of communicative competence are considered, it is clear that they may profitably be included in a FL teaching program whose goal is admittedly less than full native competence.  It is well to consider that if communicative competence (at some realistic level) is the desirable goal of FL study, then a so-called "speaking knowledge" is not closer to it than a "reading knowledge."  It would appear that communicative competence in reading and writing (with no comparable competence in aural skills) would in many circumstances be more desirable than a "speaking knowledge" of the audiolingual course variety that includes adequate structural manipulation but excludes practical use.  Hence an automatic preference for developing so-called "speaking skills" in many modern FL courses would seem to be arbitrary and unjustified.
 

    It will be recalled that one of the elements in the learning model outlined earlier was "perseverance," which was defined as the time the student is willing to spend in active learning.  This section will examine some of the relevant factors that enter into the term "willing" in this definition.
    There is a distinction to be drawn between "being interested" and "being motivated."  Interest usually refers to the condition where the source of the drive to study lies in the student; the latter sees the intrinsic value of the effort to be expended and the goal to be achieved.  “To motivate a student,” on the other hand, refers to a condition where it is felt that there is an absence of interest and hence the drive to study lies in some area extrinsic to the goal to be achieved.  It is often assumed that intrinsic interest is a more favorable condition for learning than supplied motivation, although the evidence on this matter is ambiguous.  Thus, Large (in Carroll, 1960) reported that, when in the depression adults were paid to follow experimental courses in Russian, they learned equally well regardless of their stated interest in the task (p. 66).  Dunkel (also in Carroll, 1960) found in a 1948 report that monetary rewards did not significantly improve performance in an artificial language learning task (p.66).  Carroll himself reports that a person's likes or dislikes for FL study were unrelated to aptitude or achievement (1960, p. 66).  He concludes: “From these results one may infer that as long as learners remain cooperative and actively engage in learning whether they want to or not, motivational differences will not make much difference in achievement.  Motivation will be related to achievement only when it affects how well students will persevere in active learning efforts in a situation in which they are relatively free to lag in attention, as in public schools” (1960, p.66).  However, it will be recalled that Pimsleur, et al. (1964), reported that their “Interest Test 1” was one of the tests on which underachievers were significantly lower than the controls, and they found it helpful to include it in their Aptitude Battery as a useful predictor of success.  Similarly, after a series of investigations on the study of French in the Montreal, Main, and Louisiana settings, Lamber (1963) reaches the following conclusion: “The results indicate that, similar to the Montreal studies, two independent factors underlie the development of skill in learning a second language: an intellectual capacity and an appropriate attitudinal orientation toward the other language group coupled with a determined motivation to learn the language" (p. 117).
    The investigations of Lambert and his associates at McGill University raise some important questions for a “social psychology of bilingualism” which are mot normally considered within the scope of FL training, yet indicate that the study of an FL may have some psychological consequences not unlike those experienced by immigrants in their efforts of acculturation.  The individual's reactions to these psychologically significant processes are believed to affect the learner’s motivation for study and, hence, his success in achievement.  Lambert puts it this way:
 
This theory, in brief, holds that an individual successfully acquiring a second language gradually adopts various aspects of behavior which characterize members of another linguistic-cultural group.  The learner’s ethnocentric tendencies and his attitudes toward the other group are believed to determine his success in learning the new language.  His motivation to learn is thought to be determined by his attitudes and by his orientation toward learning a second language.  The orientation is ‘instrumental’ in form if the purposes of language study reflect the more utilitarian value of linguistic achievement, such as getting ahead in ones occupation, and is ‘integrative’ if the student is oriiented to learn more about the other cultural community as if he desired to become a potential member of the other group.  It is also argued that  some may be anxious to learn another language as a means of being accepted in another cultural group because of dissatisfactions experienced in their own culture while other individuals may be equally as interested in another culture as they are in their own.  However., the more proficient one becomes in a second language, the more he may find that his place in his original membership group is modified at the same time as the other linguisitic-cultural group becomes something more than a reference group for him.  It may in fact become a second membership group for him.  Depending upon the compatibility of the two cultures, he may experience feelings of chagrin or regret as he loses ties in one group, mixed with the fearful anticipation of enter a relatively new group.  The concept of ‘anomie’… refers to the feelings of social uncertainty which sometimes characterize not only the bilingual by also the serious student of second language.  (1963, p. 114)

It can be seen that the question of motivation in FL study may be very complicated factor indeed.  Lambert has found that “integratively oriented” students are more successful than “instrumental oriented” learners, and he apparently believes that the latter are not normally aware that they are "trying less hard," in indeed they are.  Carroll (1960) is of the opinion that the instrumentally oriented student in fact perseveres less at FL study, which he thinks accounts for Lambert's findings.  Evidence consonant with this interpretation is provided by Politzer (1953-54), who reports that there is a direct correlation between performance in course examinations for college students and number of hours spent in voluntary language laboratory periods (the latter presumably being an indication of intrinsic interest), while the correlation with time spent in doing homework (presumably an indication of extrinsic interest) is curvilinear (the students getting A's did the least amount of homework).
    In view of the apparent importance of the learner's interest in the subject being studied, it is necessary to examine the evidence on this matter.  Only two studies were found which report college student's interest in FL study.  One was carried out at Harvard in the early 1950's by Politzer (1953-54), the other at the University of Illinois in 1968 by the Liberal Arts and Sciences Students Council (unpublished).  The situation at both institutions was similar in that the students were enrolled in the FL courses for the purpose of fulfilling the FL requirement.
    Two questions in the Politzer survey have direct relevance to student interest in FL study.  The first deals with the student's selection of a particular language for study; the following options were given in he question:

    (a) no particular reason
    (b) language happened to be more easily available in the secondary school or college schedule
    (c) it is easier than any other
    (d) it is more likely to be of specific use
    (e) reason for choice was a particular interest in French (or Hispanic) civilization or literature or people

The results are tabulated by type of course in which the student were enrolled:  courses lettered A and C are those with admitted emphasis on reading; courses lettered B and D lay greater emphasis on speaking.  A and B are first year courses, C and D are second year courses.  The total sample consisted of 455 students.  Of these, the majority were in the A and C courses (184 and 189, respectively, versus 49 and 33 for the B and D courses, respectively).  The figures in the table refer to percentages.

     The answers in this table are lumped by category of response indicating clearly extrinsic interest (a, b, and c) versus probable intrinsic interest (d and e).  The following points are evident upon inspection:  (1) the students who enroll in a course with a greater emphasis on speaking (B, D) have a more intrinsic interest in the language they are studying; (2) the extent of intrinsic interest held in the first year course decreases in the second year course; (3) anywhere from 1/3 to 1/2 of the students in the A and C courses have no intrinsic interest in the language they are studying.  In addition, a breakdown of answers by grades indicates that more of the good students (in all courses) have an intrinsic interest (75%) than the bad students (53%), as indicated by their course grades.
    The second question asked the students to indicate their opinion on what they felt ought to be the primary purpose of instruction.  The choices given were:

(a) acquisition of reading knowledge
(b) ability to speak with some fluency on everyday topics
(c) acquaintance with some major literary works in the original
(d) better understanding of French (or Hispanic) culture

    To the question of what are the goals or benefits of FL study one usually obtains answers that fall into four categories (not necessarily independent):  a reading knowledge, a speaking knowledge, communicative competence, and cultural awareness.  The traditional goal of FL training and the one still dominant in the minds of undergraduate and graduate university students is the attainment of a reading capacity sufficient to serve library research requirements and to provide the possibility of enjoying major literary works in a language other than one's own.  The major goal of the Intensive Foreign Language Program which sparked the "new" audiolingual approach during World War II was to produce a "practical speaking knowledge" in the shortest possible time.  This result satisfied the immediate requirements of a need at hand: to train military, paramilitary, and diplomatic personnel assigned to overseas duty.  As a result of these efforts, coupled by the increasing number of private American citizens who travel abroad, there has developed a strong feeling among many individuals connected with education, that the primary goal of FL study in the school ought to be the development of oral skills.  With respect to this change of emphasis, Moulton raises certain important questions:

As part of a liberal education, however, we may also want the student to retain an understanding of the structure of the foreign language, just as we want him to gain and retain an understanding of the structure of his native language....Likewise, as part of a liberal education we must be interested not only i teaching our students to speak, but also in teaching them to say something worth listening to; and this means that part of their work must consist in  reading some of the great things which have been said in the particular language, some of its best works of literature.  (1962, p. 90)
 
While most teachers today would not dispute the fact that reading literature in an FL is beneficial, and indeed such is required in advanced language courses, yet the emphasis on oral skills in the popularized audiolingual courses shows that when study time is short and a choice has to be made, the preference is for a development of oral skills and literature reading is left largely to the student who wants to specialize in FLs.  The rationale for this preference is not obvious and the decision needs justification, for in fact it may not have any.  The fact is that the practical considerations for FL study such as travel and a job abroad are still not sufficiently important for most Americans, and given the dominance of English in the world, which is still increasing, these considerations may never by themselves fully justify the emphasis of oral skills in FL study.  The present wide insistence on a goal that is justifiable felt unwarranted by many students in our educational system may be responsible in large measure for the lack of high motivation in FL study.  The choice of which skill to emphasize in FL instruction ought to be made in response to the goal or need felt by the student, not by some arbitrary decision.  The emphasis on oral skills in the Intensive Foreign Language Program of World War II, in many current efforts such as the Peace Corps Training program, and in private FL training programs designed for business men, is a rational choice justified by the trainee's particular needs.  Such a single purpose goal for FL training in our schools is not similarly responsive to the potentially variable needs of the general student population.
    Irrespective of whether the goal of FL study is specified as a reading knowledge or a speaking knowledge, or both, there is one ultimate benefit that receives universal consensus.  This is that FL study increases cultural awareness, reduces ethnocentrism, and is an effective "antidote to cultural myopia."  Despite the undisputed status of this benefit, FL teachers often are incapable of making an efficient case for it to the student who often remains skeptical.  It may be instructive to examine, in some detail, the arguments that some writers have used relative to this question.
    The relationship between language and culture is often discussed in the literature in relation to the so-called linguistic or cultural relativity hypothesis, whose best known modern proponents have been Sapir and Whorf (sometimes also called the "Whorfian hypotheses" or the "Sapir-Whorf hypotheses").  Writers have disagreed concerning the strength of the influence of the language system upon thought, perception, and motivation.  The strong formulation of the hypothesis posits that with each different language system there is correlated a unique thought pattern which determines the speaker's world view.  The consequence of this position is the denial of the possibility of an exact correspondence between tow linguistic expressions that belong to different language systems.  The weak formulation only admits that coding systems in general, of which language represents one example, facilitates or inhibits memory functions.  This position denies a direct influence of the language system on perception, and admits the possibility of "equivalent expressions" across different language systems.
    Carroll has edited a book on some selected writings of Benjamin Lee Whorf (1956), but he seems to reject the strong version of the relativity hypothesis.  He inclines toward the weaker proposition that language "predisposes" an individual toward a particular view insofar as the grammatical and semantic classes peculiar to a language emphasize certain aspects of experience and the environment in a way that is different from another language.  He writes:  "Insofar as language differ in the ways they encode objective experience, language users tend to sort out and distinguish experiences differently according to the categories provided by their respective languages.  These cognitions will tend to have certain effects on behavior" (Carroll, 1963b, p. 12).  Some examples will illustrate the process Carroll has in mind.  It is well known that the Eskimos have some dozen different words for varieties of snow, and so a person who is to communicate in their language must change his usual perceptions of snowflakes by learning certain differentiations he failed to make before.  The realization that classification of objects as expressed by the names we give them is arbitrary may come when an English speaker learns Chinese and discovers that two categories of things that he has hitherto considered very different ("fruits" and "nuts") are in this language lumped into one class having but one name.  To give an example that might be more familiar, when an American learns German his attention is drawn to the fact that German speaking people must specify the manner of transportation when they talk about displacing themselves: A German cannot just "go" to his country house; he must either walk (gehen), use a vehicle (fahren), ride on horseback (reiten), or whatever.  Where English uses the verb "to pick up" something, Navaho uses several variants depending on the shape of the object that is handled-a difference which may account for the fact that Navaho speaking children pay attention to the form of objects at an earlier age than American children (Carroll, 1963b, p. 16).  An American learning Spanish must consciously make a type of distinction about the concept of "to be" (ser versue estar) which he normally doesn't think of when speaking English.  The student of a second language becomes sensitized to the fact  that many of his values ar not shared by members of another culture when he learns the different connotations of so-called "translation equivalent" words:  The word "fat" in American English has a negative connotation, while in the Hindi language, among others, it carries a positive effect.  It is sometimes argued that this kind of "cultural awareness" could be taught in an anthropology type course and one need not go to the trouble of acquiring a particular FL.  Actually this is highly unlikely for it assumes that all or most cultural differences can be explicitly identified and stated verbally-a claim that no serious student of anthropology would dare make.  In fact, recent developments in anthropology use a type of linguistic contrastive analysis to isolate cultural differences ("componential analysis"),  and in psychology methods of "comparative psycholinguistics" are employed to isolate cross-cultural differences of a psychological nature (Jakobovits, 1966).  There is recent evidence that linguistic differences in the connotation of translating equivalent words provides a clue to deep psychological factors related to the need system (personality) of individuals that distinguish economically "advanced" from economically "disadvantaged societies (Jakobovits, 196a).  One recalls in this context Lambert's observation that becoming bilingual is more than just acquiring a "third signaling system": it may have weighty consequences not unlike the acculturation process of the immigrant.  Ervin (1955) has shown that the personality profile of a bilingual as revealed by his responses to the T.A.T.¹ picture cards varies depending on which of the two languages he uses.  In extreme instances, one may even speak of "bilingual schizophrenia" (Jakobovits, 1967b).
    Of course, one would not expect such strong effects to take place as a result of FL study in the school situation.  Bit it is a question of degree only, and Carroll's following remark is expected to apply, to some extent, to all students engaged in FL study:  "An individual learning a second language must be taught to observe and codify experience as nearly as possible in the same way as native speakers of that language" (1963b, p. 17).  It is in this, much more than in his pronunciation, that the native speaker becomes a worthwhile object to be modeled.
 

1.  Thematic Apperception Tests
 

    The main conclusions that emerge from a review of the research on language learning and teaching as it is relevant to an evaluation of the FL requirement will now be summarized.

    1.  Two major approaches in method of instructing FLs can be identified: the habit-skill approach
         that emphasizes oral skills through a method of repetitive pattern drills and the rule-governed
         grammar approach that emphasizes the knowledge of structure.  Both approaches draw upon
         modern linguistic theory, but the extent to which reading and writing skills are also emphasized
         varies greatly within either approach.
    2.  The habit-skill approach is based on a theoretically untenable position and the justification
         given for repetitive pattern drills appears invalid.
    3.  A scientific assessment of effectiveness of different methods of instruction is not now possible.
         Comparative studies that have been carried out support the following generalization: the
         student learns, if anything, precisely what he is being taught and there are no mysterious
         transfer effects across different language skills.
    4.  A proposal to give FL credit on the basis of attained proficiency in self-instructional
         programs for some students with high FL aptitude appears to be feasible.
    5.  The quality of instruction depends on the extent to which it is made clear to the learner just
         what he supposed to be learning at each level.  Hence, a "standard" method of instruction
         which does not vary with the learner's characteristics (aptitude and ability to understand
         instructions) cannot be of high quality.
    6.  Evidence is now available that an important factor determining success in FL study is a special
         talent for languages and that this FL aptitude can be estimated by a short test.
    7.  Given a certain known level of FL aptitude, one can specify how long it will take to attain a
         particular level of proficiency.
    8.  Given a certain limited amount of time available for study, one can predict what proportion of
         students will fail to attain a certain level of proficiency on the basis of their estimated FL
         aptitude score.
    9.  Phonetic discrimination per se (i.e., hearing ability) is not an important factor in FL aptitude.
         What is crucial is the ability to code phonetic material so that it can be stored in memory.  In
         all, four separate abilities have been identified as components of FL aptitude.
  10.  Existing FL aptitude tests can be used for placement and diagnostic purposes.  When they
         are combined with a trial course, it should be possible to predict the level of attainment for
         each of the separate language skills.  These expectations could then be used to establish
         variable FL requirements for each student.
  11.  Standard proficiency tests are inadequate on two accounts: (a) they do not indicate the
         student's absolute competence-only his relative standing to a "typical group"; (b) they are
         intended to test mastery of mechanical manipulation or structural knowledge and say nothing
         about communicative competence.  What is needed is a specification of levels of standards in
         terms of practical requirements in communication.
  12.  Attainment in FL study can be evaluated on two levels: degree of mastery of mechanical
         manipulation and extent of communicative competence.  There appears to be no good reason
         or withholding training in the latter until advanced mastery of the former.  Teaching at both
         levels conjointly would be preferable on several grounds.
  13.  As it is usually defined in FL courses or requirements, a "speaking knowledge" is not closer to
         communicative competence than a "reading knowledge."  The emphasis of the former over the
          latter does now appear to be justified.
  14.  A consideration of motivational and attitudinal factors in FL study is relevant from two points
         of view: (a) the way in which these affect learners' perseverance (an intrinsic-integrative
         orientation has been found to be superior to an extrinsic-instrumental orientation); (b) the way
          in which these affect the individual's reactions to contact with a foreign culture (there is
         evidence that becoming bilingual carries with it the tendency of becoming bicultural).
  15.  Two available surveys of college students' interest in FL study indicate that (a) one-third to
         one-half have no intrinsic interest in the language they are taking; (b) one-half consider the
         primary goal to be the development of reading knowledge of the language; and (c) most of
         them disapprove of the college graduation FL requirement and almost one-half actually feel
         that FL study has been detrimental to them.
 16.  Given the wide distribution of FL aptitude scores in the general student population, it is
        unrealistic to expect a uniform level of achievement.  The FL teacher must adjust his
        expectation of achievement to be attained to the student's particular aptitude
 17.  The low level of intrinsic interest in FL study on the part of many students may be a result of a
        number of undesirable features of present practices in FL instruction.  Some of these can be
        mentioned: (a) a failure to clarify appropriate criteria for student self-evaluation of his progress;
        (b) insistence on a degree of phonetic mastery that is not justified by the requirements of
        communication; (c) inadequate counseling on the choice of an FL in terms of considerations
        appropriate to the student's perceived needs' and (d) failure to make FL study seem relevant
        to the student in terms of its practical use.
 18.  A single-purpose goal for FL study, such as the development of oral skills in the "progressive"
        audiolingual courses, is not responsive to the potentially variable needs of the general student
        population.
 19.  Although proponents of FL study agree that one of its main benefits is increased cultural
        awareness, they have apparently failed to communicate this idea to many students.  It is
        possible to justify this claim on a number of grounds and it behooves the FL teacher to try to
        make a better case for it.
 

    1.  It is my opinion that most college students who successfully meet present requirements fail to
         achieve a level of proficiency that allows them to reap the benefits of FL study, whether it be
         at the humanistic level or the practical level of communicative competence.
    2.  I believe that this failure of the intended goal has several causes, the most important one
         being the following: (a) FL study is not perceived by the students as relevant to their
         educational needs and aspirations,  This feeling is strengthened by blanket college
         requirements which they view as archaic, arbitrary, and insensitive to their wishes; (b) serious
         benefits can be derived, entangles the student in a psychological involvement that may lead
         him to invidious comparisons between the foreign culture and his own.  To some individuals,
         this involvement may be threatening, especially when he feels that it is being "forced down his
         throat."  Failure to achieve any meaningful proficiency is an effective protection against such
         perceived threat; (c) present methods of instruction are geared toward a standard goal, while
         the students' needs are variable.  This helps reinforce the feeling of the irrelevance of FL
         study.  Apart from neglecting individual needs, such instruction disregards the variation that
         exists in FL aptitude.
    3.  The acquisition of a meaningful level of language competence cannot be achieved without
         intrinsic motivation on the part of the learner.  No amount of duress in the form of a barrier
         against graduation can change this.  In this sense, the FL requirement is self-defeating.
 
    Given these considerations, I would like to suggest some alternatives, which in my opinion, will
be more effective in promoting the goal that the present requirement attempts, but fails, to promote.

    1.  The student ought to be given as much latitude as he feels he needs in the following areas: (a)
         the choice of a language (he ought not to be pressured to "stick to" a language he took in high
         school at a time when his needs may have been different and when the choices available to
         him were restricted); (b) the type of instruction and the skills to be emphasized (he ought not
         to be made to suffer in self-respect by choosing a "starred"² course when his interests lie in
         developing reading and writing skills; typically, starred courses are for "D" students).  The high
         aptitude student ought to be given the opportunity of pursuing self-instructional programs
         where he can achieve a desired level of proficiency at a rate much faster than the typical
         course affords; and (c) the amount of credit he wishes to receive in FL study.  This last point
         is elaborated in the following recommendation.

² A term used at the Univ. of Illinois to denote sections for individuals with special problems.

Korean                   2                  4             6               15               16                 24*         30*

Japanese                 3                 3              4               12               15                 18           24
^{____________________________________________________________________________________}
Arabic                    1                 6              9               No              No                No           No

Vietnamese             2                 4              6               12               15                 18*          24*

Thai                        3                 3              4                9                12                 15*          18*
^{____________________________________________________________________________________}
Notes:  (a) "No" entries indicate that it is not practical to achieve that level of proficiency on a one-hour-a-day
                      basis.
                (b) Entries with an asterisk indicate that one must add three months in part-time training and using the
                      language, preferably in the field.
                (c) For definitions of levels and hours, see text, p. 441.
 

    2.  A variable credit allotment ought to be made available based on the student's attainment of certain defined levels of standards stated in terms relevant to competence in language use.  The following plan is intended as an illustration rather than a firm proposal; (See Table 2)
    The scheme of variable credit illustrated here is flexible enough to meet the variable needs and interests of the student body.  Credit can be obtained at any level with three types of skills.  An ambitious student with high aptitude can pile up as many as 18 1/2 credit units with the combination [W₆ + O₃ + H_L + H_C].  Or, a student may simply choose to write a technical paper on his major subject and receive  4 1/2 credit units for W₆, etc.  The exact credit allotments and levels of standards should be worked out more carefully and with some justifiable rationale.
    3.  I would also recommend that each department in a college spell out for its majors the desired levels of standards, types of skills that it recommends, and in which language(s).  These recommendations ought to be flexible and be based on demonstrable usefulness to the program in question.
    It is my belief that if these recommendations are put into effect, FL study will quickly loose its notorious status as "the bad child" of the college curriculum and will provide the type of benefits that we all want the liberally educated person to experience.