Those were the Good Ol’ Days: Historical Trends in Language Sampling and Analysis

Rhea Paul, PhD, CCC-SLP Sacred Heart University Rhea received her Ph.D. from the University of Wisconsin-Madison in 1981, and has taught language development and disorders in children for over 30 years. You probably know her as the author of the quintessential textbook: Language Disorders from Infancy Through Adolescence. Her latest projects include her podcast, Let’s Talk About Super Special Kids & Cake, with co-host Donia Fahim. Check it out on Apple Podcasts or Spotify: Apple Podcasts: https://podcasts.apple.com/us/podcast/lets-talk-about-super-special-kids-cake/id1446307245 Spotify: https://open.spotify.com/show/1I7FfBEWww2cJgtnJKoCGi

Do you remember 8 track tapes? Listening to them on your car’s tape deck or on a boom box? If that seems like ancient history to you, or if you don’t remember it at all, you may never have thought about what life was like for speech-language pathologists (SLPs) before we could record whatever we like on our phones. But in point of fact, people interested in children’s language development and disorders have been recording kids’ speech with whatever technology was available for well over 100 years. And there were quite a lot of them! One was Charles Darwin, who kept detailed handwritten recordings of his children’s early speech productions in the 19^th century (Bar-Adon & Leopold, 1971). If you are an SLP, you may have come across the name Mildred Templin (1957) in your studies. She used language sampling to establish norms for the acquisition of speech sounds, vocabulary, and amount of talking. She was perhaps the first to note significant differences in language production based on SES. Working in the 1930s-50s, before the age of even the tape recorder, she sat in the back of classrooms writing down everything she heard children say. But the practice of language sampling was really kick-started by the advent of renewed interest in syntax as a window into the organization of language in the mind, sparked by Chomsky’s (1957) seminal work. In the 1960s, psychologists with an interest in language began looking at young children’s spontaneous speech for information about their grammatical acquisition patterns, using language samples taken over many months from a very small number of children to trace developments such as noun and verb inflections, negatives and questions, and sentence structures. Most of these pioneers were still hand-transcribing from live observations. But another innovation that arose just around the time of Chomsky’s nudge toward interest in child language as a “natural experiment” for describing syntactic complexity, was the availability of the magnetic tape recorder, which changed everything in LSA. It was soon possible for anyone to, relatively cheaply and easily, to record audio material. MacWhinney and Snow (1984) describe the impact of these devices:

“…It became possible to make records of child language productions that were far more complete than the diaries that had been compiled from pencil and paper notes. With these new data, the field was able to advance well beyond the level of anecdotal summaries that had characterized the first 100 years of child language research. The new methodology permitted full recording of not just the child, but also those who interacted with the child. This new possibility slowly turned the field's attention to studies of the input to the language- learning child. During the 1960s and 1970s, researchers continually improved their use of recording technology, using directional microphones, wireless microphones, well-justified sampling strategies, non-intrusive recording methods, and detailed systems for coding aspects of the communicative context” (p. 272).

Even with recording devices, though, samples still needed to be transcribed and analyzed, which remained a laborious job. As these technical developments were progressing, and the theoretical interest in language development as a means of testing various hypotheses about the nature of language learning was driven by early psycholinguists, the idea that looking at children’s spontaneous language could inform clinical practice for children with speech and language disorders began to seep into the communication disorders (then called ‘speech pathology’) literature. This trend was spurred by Jon Eisenson's (1968) publication of one of the first papers in the ‘speech pathology’ literature on children’s developmental language disorders. The interest in understanding language development as a key to remediating language delays began to be integrated with the notion that a distinct disorder that primarily affected oral language was within the purview of the ‘speech therapist.’ Laura Lee and her colleagues (Lee, 1966; Lee & Canter, 1971) took the next step in the process of connecting the study of children’s speech output to understanding development and disorders. They used a relatively large (160) set of child speech samples to create an outline of typical developmental sequence of syntactic acquisition relative to adult targets in children from 3-7 years. They then applied this sequence to the analysis of language from children who were clinically referred for language disorders, creating perhaps the first relatively comprehensive approach to analysis of language samples. Crystal, Fletcher, & Garman (1976) introduced another procedure with a similar purpose, which was used widely in the UK and Canada at around the time the DSS. Jon Miller (1981) was one of the first in the field of Communication Disorders to advocate for using language samples as a source of data for looking across a range of language domains, including not only syntax, but also phonology, vocabulary and pragmatics.  He argued that just as children often make different errors in single word articulation tests than they do in running speech (Faircloth, & Faircloth, 1970), children also make different errors in sentence imitation tasks than they do in authentic conversation. Miller asserted that the best guide to the real difficulties children with language disorders face in expressive language is the speech they present in authentic interactions, which should be studied and analyzed in order to establish the most ecologically valid targets for intervention. Language sampling became one tool in the toolbox of clinicians working with children with language disabilities. Still, even with recording equipment, LSA required recordings to be collected, transcribed (still a laborious process), then analyzed in some way. As the 1980s progressed, two things became clear. First, it was VERY hard to get clinicians to collect, transcribe and analyze language samples in the course of their clinical work. Even though this process did not necessarily take longer than giving a comprehensive standardized test, such as the ITPA or PLS, clinicians balked, perhaps at the learning curve more than the time. At the same time, personal computers were becoming more ubiquitous, and several groups began enlisting the efficiencies computers could offer to the task of LSA. MacWhinney & Snow (1985) were perhaps the first to make the connection between LSA and the personal computing revolution. They obtained funding from the MacArthur Foundation to support the development of a child language data exchange system (CHILDES), publishing their paper to invite all researchers to contribute transcribed language samples to the database, laid out rules for transcription, coding and data sharing, and began the process of developing analysis routines that would, over the years be automated. At about the same time, Miller & Chapman (1985) developed the Systematic Analysis of Language Samples (SALT), an alternative system for transcription and automated analysis. The SALT system, accompanied by its Larry Shribergs’ PEPPER utility for analyzing phonological data, was developed with clinicians as well as researchers in mind, and it remains the preferred system for many in clinical practice. While none of automated innovations so far have achieved either of our holy grails: automatic transcription or increased uptake by practicing clinicians of language sampling as an assessment technique, these innovations have greatly contributed to the use of LSA in research on both typical and language impaired children. As technology progresses, we can hope to see decreases in the effort needed to implement language sampling and analysis in the everyday work of the SLP. _______________ References Bar-Adon, A., & Leopold, W. F. (1971). Child Language; A Book of Readings. Englewood Cliffs, NJ: Prentice-Hall. Bloom, L., (1970). Language Development: Form and Function in Emerging Grammars. Cambridge, Mass.: MIT. Brown, R., (1968). The development of wh-questions in child speech. Journal of verbal learning and verbal Behavior, 7, 279-290. Brown, R., & Fraser, C. (1964). The acquisition of syntax. Child Developm. Monogr., 29, 43-79 Cazden,C., (1965). Environmental assistance to the child's acquisition of grammar. Unpublished doctoral dissertation, Harvard University. Cazden,C., (1968). The acquisition of noun and verb inflections. Child Development, 39, 433-448. Chamberlain, A. F. & Chamberlain, J. (1904). Studies of a child. Pedagogical Seminary. Taylor & Francis. Chomsky, N. (1957). Syntactic Structures. The Hague: Mouton. Crystal, D., Fletcher, P., and Garman, M. The grammatical analysis of language disability: a procedure for assessment and remediation. London: Edward Arnold, 1976. Eisenson, J. (1968). Developmental aphasia: A speculative view with therapeutic implications. Journal of Speech and Hearing Disorders, 33, 3-13 Faircloth M., & Faircloth S. (1970). An analysis of the articulatory behavior of a speech-defective child in connected speech and in isolated-word responses. Journal of Speech and Hearing Disorders, 35(1), 51–61. Guillaume, P. (1927). Imitation in children. (Translated by E. P. Halperin.) Chicago: University of Chicago Press. Klima, E., and Bellugi, U., (1966). Syntactic regularities in the speech of children. In J. Lyons and R. J. Wales (Eds.), Psycholinguistics Papers. Edinburgh: Edinburgh Univ. Lee, L. (1966). Developmental sentence types: A method for comparing normal and deviant syntactic development. J. Speech Hearing Dis., 31, 311-330. Lee L., & Canter S. (1971). Developmental Sentence Scoring: A Clinical procedure for estimating syntactic development in children’s spontaneous speech. Journal of Speech, Language, and Hearing Research, 36, 315-340. Leopold, W. K. (1939). Speech development of a bilingual child. Evanston: Northwestern University Press. Lewis, M. M. (1951). Infant speech (2nd ed.). London: Routledge & Kegan Paul. Long, S. H., Fey, M. E., & Channell, R. W. (1993). Computerized profiling. San Antonio, TX: The Psychological Corporation. MacWhinney, B., & Snow, C. (1985). The child language data exchange system*. Journal of Child Language, 12(2), 271–295. Menyuk, P., (1969). Sentences children use. Cambridge, Mass.: MIT (1969). Miller, J. F. (1981). Assessing language production in children: Experimental procedures. Univ Park Press. Miller, J., & Chapman, R. (1985). Systematic analysis of language transcripts (SALT). Madison: University of Wisconsin. Miller, W., and Ervin, S., (1964). The development of grammar in child language. Child Development Monograph, 29, 9-34. Moore, K. C. (1896). Mental development of a child. Psychological Review (monograph supplement) no. 3. Pye, C. (1987). Pye Analysis of Language. Lawrence, KS: University of Kansas. Templin, M. C. (1957). Certain language skills in children; their development and interrelationships. Minneapolis, MN, US: University of Minnesota Press. Weiner, F. F. (1986). Parrot easy language sample analysis (PELSA)[Computer software]. State College, PA: Parrot Software.