Purpose: We show political divisions in perceptions of police officers even before the divisive political and social events of 2016. We do so using respondents' interpretations of surprising and ambiguous headlines involving police officers (e.g., assumptions about what happened or who was involved).Methodology/Approach: We use affect control theory's ABO event structure and derivations of this structure to construct a set of headlines that describe ostensibly good people (A) doing bad things (B) to other good people (O) or are ambiguous on one or more of these components. We present 517 MTurk respondents with a set of 7 headlines and collect quantitative and qualitative data on their reactions to, and interpretations of, these headline events.Findings: Police headlines generate interest among readers. When interpreting events, respondents are less likely to modify or redefine police officers compared to other actors However, assumptions related to ambiguous events involving police differ by political orientation. Liberals view police more negatively than conservatives, in part because they imagine them doing worse things to slightly better people. Qualitative analyses support and shed light on the mechanisms underlying this and other partisan effects.Research Limitations: The research was designed to examine interest in headline structure, not specific actors. Thus, the patterns unique to police and political differences were not an original focus. We believe these inductive results are informative, but a study expressly designed to test hypotheses regarding perceptions of events with police officers is recommended for future work.Practical and Social Implications: Understanding the political divide in perceptions of police and the potential of media coverage for exacerbating these effects is essential and related to ACTs growing interest in meaning divergence.