Issues in Subject Analysis and Description of Political Cartoons

Abstract

Political cartoons are not meant to record visual evidence of an event as a photo might, neither are they created to act as icons for the events that they speak to. Rather, they treat the events of their day with an acknowledged slant in the point-of-view, draw correlations between events when such correlations might exist only in the mind of the artist, or deride (or, rarely, admire) individuals or organizations. In all cases, political cartoons fall far more on Fidelâ s Object pole than they do on her Data pole (1997). Indexing political cartoons offers a unique challenge in the larger realm of indexing images. But while subject access has been the focus of image indexing research in recent years, and is a robust and active topic of discussion and debate, it has rarely been turned to the realm of indexing opinion, visual or otherwise. Will Armitage and Enserâ s Panofsky-Shatford mode/facet matrix (1997) be more useful in such work than Jorgensenâ s 12 classes (1994), or will an entirely new measure of subject need to be developed? This paper asks questions within this realm of image indexing as it pertains to political cartoons. Do the subject indexing needs of political cartoons change if the focus is a social or historical issue rather than a political one? There are times when a political cartoonist will choose to address an issue that has less to do with political action or intrigue and more to do with a social ill or trend (such as teen pregnancy), or a historical event of a limited nature (such as the death of a statesman). Does this focus change the needs and requirements of providing subject access to cartoons from that needed for images of a more political nature? Is this a part of the â biographyâ of an image (Roberts, 2001)? Is the event (or events) that inspired a cartoon the de facto subject(s) of it? Proceeding from the proposition that some sort of newsworthy events are the progenitors of political cartoons, do those events serve as the subject elements in cartoon indexing schemes? Is Svenoniousâ subject/predicate model (1994) complete enough? Is the subject of a cartoon determined by what must be known to â get itâ ? Can a cartoon have more than two subjects? Some cartoons make their point though the juxtaposition of disparate items. If a cartoon uses two disparate news items to make a point, are there two subjects? If so, is there a hierarchical relationship â superordinate/subordinate, major/minor, primary/secondary â between them? Is Mai (2005) correct in saying that the meaning of a document changes, depending on the question being asked of it? On the opposite end of the spectrum, can a cartoon refer to something that is not useful as a subject descriptor?