Counting Affixes with Morpholex: A Response to McLean and Stoeckel (2021)

Autor/in	Cobb, Tom
Titel	Counting Affixes with Morpholex: A Response to McLean and Stoeckel (2021)
Quelle	In: Reading in a Foreign Language, 34 (2022) 1, S.165-171 (7 Seiten) PDF als Volltext Verfügbarkeit
Sprache	englisch
Dokumenttyp	gedruckt; online; Zeitschriftenaufsatz
Schlagwörter	Morphemes; English (Second Language); Second Language Learning; Second Language Instruction; Reading Processes; Morphology (Languages); Reading Comprehension; Barriers; Computational Linguistics; Reading Research; Criticism; Vocabulary Skills; Reading Difficulties + Suchen Sie Ihr Suchwort? Morphem; English as second language; English; Second Language; Englisch als Zweitsprache; Zweitsprachenerwerb; Fremdsprachenunterricht; Leseprozess; Morphology; Morphologie; Leseverstehen; Linguistics; Computerlinguistik; Leseforschung; Kritik; Aktiver Wortschatz; Reading difficulty; Leseschwierigkeit
Abstract	Some words in English are derived words, i.e., words with an affix that changes their part of speech, like know "(v)" [right arrow] "knowable (adj)," or meaning, like "pleasant (adj)" [right arrow] "unpleasant (adj)." As a reading teacher, Tom Cobb had always believed that derived words were not particularly problematic for his English as a second language (ESL) and foreign language (EFL) learners, owing to the recurrence of a relatively small number of different derivational affixes in what they were reading. Their problem seemed rather not knowing enough words in any form. The opposing belief, held by researchers like McLean (2017), is that the roughly 120 derivational affixes of English are inscrutable to their learners, even if they know the base words they are attached to, and thus present a significant obstacle to reading comprehension. Such a disparity of beliefs, held by people with similar experience of learners, texts, and affixes and a professional interest in the matter, shows the need for an objective way of counting affixes such as text analysis computer software might contribute to. Laufer and Cobb (2020) conducted an empirical investigation into this question, for which I developed Morpholex v.3.1, a computer program that assesses the morphological load of an entered text. The output has two main parts, the proportion of base, inflected, and derived words in a text, and the cumulative token coverage each provides. Morpholex v.3.2 was recently reviewed in "Reading in a Foreign Language" by McLean and Stoeckel (2021) as part of a response to Laufer (2021) in the same volume. The review raised some useful points, chief among them that Morpholex underestimates the number of derivational morphemes in a given text by counting only once words that in fact bear two affixes, whether an inflectional and a derivational (like "teachers" [right arrow "teach" + "er" + "s," counted only as an inflection while it is also a derivation) or two derivational (like "remorselessly" [right arrow] "remorse" + "less" + "ly," counted only as one derivation rather than two). The criticism is legitimate; it was based on Laufer and Cobb's decision to put program development on hold while the paper was in review and in press. It is now more than a year since publication, during which a follow-up piece has been produced (Cobb & Laufer, 2021), Morpholex has collected megabytes of user data, and program development has resumed--partly thanks to nudges from McLean and Stoeckel and other critics. [For McLean and Stoeckel's 2021 article, "Lexical Mastery Thresholds and Lexical Units: A Reply to Laufer," see EJ1316857. For Laufer's 2021 article, "Lexical Thresholds and Alleged Threats to Validity: A Storm in a Teacup?," see EJ1317203.] (ERIC).
Anmerkungen	National Foreign Language Resource Center at University of Hawaii. 1859 East-West Road #106, Honolulu, HI 96822. e-mail: readfl@hawaii.edu; Web site: https://nflrc.hawaii.edu/rfl/
Erfasst von	ERIC (Education Resources Information Center), Washington, DC
Update	2024/1/01