The current study is designed to investigate the prosodic patterns of English sentences consisting of content words. English, as a stress-timed language, favors a

rhythmic pattern of strong-weak alternations. That is, stressed and unstressed syllables appear regularly, generating eurhythmic patterns. This kind of rhythm principle

sometimes conflicts with another principle that content words are usually stressed and function words are not, especially in utterances composed of content words alone. An experiment was conducted to investigate whether English native speakers assign stress on all content words to convey new information that the content words entail or create rhythmic patterns, giving up semantic informativeness to some degree. The results showed that the native speakers did not produce every content word with pitch accents. Approximately 50% of the utterances were produced with pitch accents on every two or three words. Another 20% of the data were uttered with controlled pitch patterns (downstepped high pitch accents or intervening low phrasal accents) in order to avoid continuous strong beats. Thus, more than 70% of the total recordings showed rhythmic patterns at the utterance level, generating accentual feet where strong (accented) and weak (unaccented) words alternate. (University of Seoul ⋅ Dongduk Women’s University)