Vocabulary in Primary School Tamil Textbooks A CorpusBased Analysis

Understanding how vocabulary is presented in earlygrade Tamil textbooks is crucial for educators, curriculum designers, and languagepolicy makers. This page summarises a corpusbased investigation that examined the lexical characteristics of the Tamil textbooks used in Grades15 across the State Board curriculum (20202024 editions). The analysis focuses on three core issues:

1. Quantitative profile of the lexical inventory (typetoken ratio, wordfrequency distribution, and lexical density).
2. Qualitative patterns (semantic fields, cognate density, and morphological complexity).
3. Pedagogical implications and recommendations for vocabulary enrichment.

1. Methodology

The study employed a mixedmethod approach:

• Corpus compilation: All narrative, prose, and poem sections from the five grades were digitised, yielding 3226000 tokens ( 645000 words per grade).
• Cleaning & tagging: NonTamil symbols, page numbers, and English insertions were removed. The Tamil POS tagger from the TamilPOS project was used for lemmatisation.
• Statistical analysis: TypeToken Ratio (TTR), HapaxLegomena proportion, Zipfs law plots, and lexical density calculations were performed with the R packages quanteda and tidytext.
• Semantic categorisation: WordNetTamil and manual annotation identified semantic fields (e.g., kinship, nature, school life).

2. Quantitative Findings

2.1 Vocabulary Size and Growth

The cumulative type count across the five grades is 28714 distinct lemmas. Growth is roughly linear:

• Grade1 4322 types (TTR0.067)
• Grade2 5108 types (TTR0.064)
• Grade3 6031 types (TTR0.060)
• Grade4 6874 types (TTR0.057)
• Grade5 6379 types (TTR0.054)

Although the number of types increases, the TTR declines, reflecting a natural shift from highly repetitive to more contentrich texts.

2.2 Frequency Distribution

Zipfian analysis shows a steep slope (1.12), indicating a strong concentration of highfrequency items. The 100 most frequent lemmas account for 32% of all tokens, and the top 1000 for 68%.

2.3 Lexical Density

Lexical density (ratio of content words to total words) rises from 39% in Grade1 to 46% in Grade5. This trend mirrors the curriculums aim to move from simple everyday talk to more abstract academic discourse.

3. Qualitative Observations

3.1 Dominant Semantic Fields

Analysis of the 2000 most frequent lemmas reveals the following dominant fields:

These fields align with the socialcultural goals of primary education but also indicate limited exposure to scientific or technological terminology.

3.2 Cognates and Loanwords

Approximately 7% of the distinct lemmas are Sanskritderived cognates, while English loanwords make up 2%. The presence of Sanskrit items is expected, given Tamils classical lexicon, yet the low English proportion suggests minimal early exposure to global scientific vocabulary.

3.3 Morphological Complexity

Tamils agglutinative nature results in a high proportion of inflected forms. The corpus shows:

• Verb forms: 42% of tokens are verb inflections (tense, mood, aspect).
• Nominal suffixes (case, number, honorific): 35% of tokens.
• Compound words (sandhi) account for 6% of types, indicating moderate lexical productivity.

4. Pedagogical Implications

1. Balanced Lexical Exposure: While everyday vocabulary is wellrepresented, the textbooks lack systematic introduction of scientific and technological terms. Supplementary reading lists or thematic units could address this gap.
2. Explicit Morphology Instruction: The high frequency of inflected forms suggests that learners encounter morphological complexity early. Targeted exercises on case markers and verb conjugation can enhance reading comprehension.
3. Vocabulary Recycling: Highfrequency items appear repeatedly across grades, which supports reinforcement. However, many lowerfrequency words (hapaxlegomena) appear only once, limiting chances for consolidation. Recurring use of these words in later grades would aid retention.
4. Semantic Field Expansion: Introducing new semantic fields gradually (e.g., health, technology, civic education) can broaden childrens world knowledge without overloading them.
5. Use of Cognates: The presence of Sanskrit cognates can be leveraged for etymological awareness, fostering metalinguistic insight. Conversely, English loanwords, though few, should be presented with clear Tamil equivalents to avoid lexical ambiguity.

5. Recommendations for Curriculum Designers

• Develop a corevocabulary list of 2000 lemmas that appear across all grades, ensuring that each is encountered at least three times.
• Integrate a lexical enrichment module in Grades45 that introduces 300400 new academic words (science, maths, social studies) with contextual illustrations.
• Adopt a morphologyfocused worksheet series that isolates case endings and verb suffixes for practice.
• Include a semanticfield map at the end of each textbook chapter that visualises how new words relate to previously learned domains.
• Provide teachers with a digital lexicon tool that shows frequency, morphological variants, and usage examples drawn directly from the corpus.

6. Conclusion

The corpusbased analysis demonstrates that primaryschool Tamil textbooks present a solid foundation of everyday vocabulary, strong morphological exposure, and a clear semantic focus on family, nature, and school life. Nevertheless, the lexical profile reveals a need for greater diversity in academic domains and more systematic recycling of lowfrequency words. By adopting the suggested pedagogical strategies and curriculum adjustments, educators can promote a richer, more balanced vocabulary development that prepares students for later academic challenges and for active participation in a multilingual world.

Keywords: Tamil, primary education, vocabulary, corpus linguistics, lexical density, curriculum development.