Researchers investigating the incidental acquisition of vocabulary through reading have often required participants to read in controlled classroom or lab environments. This method delimits reading to short texts read in one sitting and fashions an anomalous reading context (i. e. not mimicking ecologically valid extensive reading situations). To investigate whether the physical reading context affects incidental vocabulary acquisition, an empirical study was conducted with two groups of participants that read the same 36,711-token novel containing 49 target words-in-class readers (n = 48) and out-of-class readers (n = 32). Results showed reading context has a large effect on vocabulary recall and a medium effect on vocabulary recognition. Medium correlations were found between incidental acquisition and reading context as well as second language (L2) vocabulary size. Two standard three-explanatory-variable (L2 vocabulary size, reading context, reading time) multiple regressions accounted for 40% of the variance in vocabulary recall and 44.5% of the variance in vocabulary recognition. Nuanced distinctions between in-class and out-of-class readers were uncovered by analyzing responses to open-ended reflective questionnaire items about the study, novel, target vocabulary, and vocabulary learning strategies. A lens of criticality was used to discuss the findings in terms of their pedagogical and methodological implications.