The Teacher's Word Book

Thorndike argues that vocabulary instruction should be based on the actual frequency with which words occur in the English language, rather than on tradition, intuition, or the assumption that literary words are inherently more valuable than common ones. The book rejects the practice of selecting vocabulary for instruction on the basis of what teachers or textbook authors happen to consider important. Instead, Thorndike contends that knowing which words are most common allows teachers to sequence instruction efficiently, ensuring that students master the vocabulary they will encounter most often in reading. This empirical approach to vocabulary was a direct application of his conviction that educational practice should be grounded in measurement and the laws of learning.

Thorndike and his assistants compiled the list by counting words in a large corpus of texts representing different types of English usage. The sources included the Bible, English classics, elementary school textbooks, and other widely read materials. The total count ran to approximately 4.5 million words, though this figure was later disputed on the grounds that Thorndike claimed 4.5 million words counted but failed to define what he meant by 'word' and how he counted. Words were ranked by frequency, and the resulting list of about 10,000 words was presented with indications of their relative frequency. The book was intended as a practical reference for teachers, not a theoretical treatise, though it rested on a substantial empirical foundation.

Critics questioned Thorndike's claim to have counted 4.5 million words, noting that he did not clearly define what counted as a 'word' and that the figure seemed implausibly large given the methods available at the time. Thorndike defended the count, arguing that the number referred to the total number of word occurrences in the corpus, not the number of distinct words, and that his definition of a word was consistent with standard linguistic usage. The controversy was significant because it raised questions about the reliability of frequency counts compiled without mechanical aids. Later word frequency counts, such as those by Henry Kučera and W. Nelson Francis, used computer methods and more explicit criteria, and they largely confirmed the general shape of Thorndike's list while revising specific rankings.

The Teacher's Word Book is an application of Thorndike's connectionist principles to the practical problem of vocabulary instruction. His educational psychology had argued that learning is specific and that transfer is limited, which implied that students need to be taught the specific words they will actually encounter rather than general principles of vocabulary. The word book provided the empirical basis for deciding which words those are. It also reflected his conviction that educational decisions should be based on measurement rather than tradition. Thorndike later produced expanded editions and related word books, including The Teacher's Word Book of 20,000 Words (1931) and The Teacher's Word Book of 30,000 Words (1944, with Irving Lorge), which extended the approach to a larger corpus.

The book was widely adopted and became a standard reference for teachers, textbook authors, and curriculum designers. Its impact on reading instruction was substantial: it provided an empirical basis for selecting vocabulary in basal readers and for grading texts by difficulty. The idea that vocabulary should be sequenced by frequency became a foundational principle in reading education, influencing the development of readability formulas and leveled readers. Some critics argued that frequency alone is an insufficient criterion for vocabulary selection, since some infrequent words are conceptually important and some frequent words are already known. Nonetheless, the book established the principle that vocabulary instruction should be informed by systematic evidence about language use.

Thorndike's basic insight—that frequency of occurrence is a crucial variable in vocabulary learning and instruction—has been strongly supported by subsequent research. Corpus linguistics has confirmed that a relatively small number of words accounts for a very large proportion of all text, and that frequency is one of the best predictors of word difficulty and acquisition. Contemporary vocabulary research continues to use frequency counts, though with more sophisticated corpora and statistical methods. However, later research has also shown that frequency alone does not determine which words should be taught: factors such as morphological complexity, conceptual difficulty, and importance for academic content also matter. The Teacher's Word Book remains a historical landmark in the empirical study of vocabulary and the first major attempt to apply frequency data to educational practice.