Background: Secondary school teachers sometimes teach and assess material outside their specialisms for reasons including staff shortages or the growing popularity of the interdisciplinary courses. We hypothesised that teacher-assessors with different subject specialisms may differ in their interpretations of frequently used words in teaching and assessment, such as 'analyse' and 'evaluate'. Differing interpretations of these seemingly generic command words are worthy of investigation because they may have implications for the clarity and success of teaching and assessment. Purpose: The aim of this small-scale study was to identify and analyse teacher-assessors' interpretative differences of command words commonly used in teaching and assessment in order to draw out implications for students' learning and assessment outcomes. Method: Written definitions for five common command words (analyse, evaluate, review, synthesise and argue) were obtained from six published sources and also from nine experienced teacher-assessors in England covering seven subjects. A directed qualitative content analysis was conducted, entailing comparisons across individual definitions and across the humanities and sciences. Findings: For all five command words, definitions were found to vary in both conceptual complexity and subject specificity, sometimes within the same subject. Fundamentally different concepts were identified for 'review' (which may or may not entail making a judgement about material) and 'argue' (which may or may not focus on reasoning). This could be an important cause of confusion for some students and may limit the depth of teaching and understanding for others. Furthermore, such differences could reduce agreement among assessors and undermine the appropriateness of the inferences drawn from the outcomes of assessments. Conclusion: This small-scale investigation highlights the importance of building shared understandings of assessment criteria. Although assessor standardisation is the norm and occurs via formal procedures within large examination boards, the emphasis on it may be considerably weaker in less regulated teacher-assessor contexts. Further research could usefully focus on training and support the needs of non-specialist teachers and interdisciplinary course developers, strengthening communities of practice.