Abstract
Disagreement is essential to scientific progress. However, the extent ofdisagreement in science, its evolution over time, and the fields in which ithappens, remains largely unknown. Leveraging a massive collection of scientifictexts, we develop a cue-phrase based approach to identify instances ofdisagreement citations across more than four million scientific articles. Usingthis method, we construct an indicator of disagreement across scientific fieldsover the 2000-2015 period. In contrast with black-box text classificationmethods, our framework is transparent and easily interpretable. We reveal adisciplinary spectrum of disagreement, with higher disagreement in the socialsciences and lower disagreement in physics and mathematics. However, detaileddisciplinary analysis demonstrates heterogeneity across sub-fields, revealingthe importance of local disciplinary cultures and epistemic characteristics ofdisagreement. Paper-level analysis reveals notable episodes of disagreement inscience, and illustrates how methodological artefacts can confound analyses ofscientific texts. These findings contribute to a broader understanding ofdisagreement and establish a foundation for future research to understandingkey processes underlying scientific progress.