Media montage

What insight can we gain about North Korea in the 1948-1950 period from the media montages?

Much can be said about the overall design and layout of the newspaper. The newspaper remains, over the two-years covered in this project, text-heavy. There is no noticeable increase in the use of non-text content over time. Most issues contain only two to three pictures, and it is not uncommon to see multiple pages in an issue containing only text. This raises the question of the effectiveness of this newspaper in delivering its message to its target audience of rank-and-file soldiers, many of whom were barely literate at the time.

Also apparent are instances of censorship, which are visible even in the low-resolution versions available on this website as white boxes running across the pages. It is impossible to determine an accurate motive, but the computer-generated white boxes that cover some of the pages are clearly censorship carried out during the digitization process in North Korea. There appears to be no pattern to this, but one conjecture is that the censored areas contain information that contradict the official version of North Korean history, especially those relating to North Korean preparations for invasion of the South, a fact that has been denied staunchly by the DPRK government despite consensus among scholars who analyzed foreign diplomatic documents and documents captured during the war. Other possibilities, such as removing pictures and speeches of later purged individuals have been overruled by the fact that many appear uncensored.

Although the montage provides a quick picture of the most obvious trends and irregularities in the newspaper, it does not allow for a microanalysis. One lesson from this media montage is that this technique works best to show changes over a much longer period of time. It was not the most fruitful to analyze just two years of the newspaper, when changes in the newspaper design or non-text content were not significant. Another lesson is that there may be a limit to a wholly visual analysis of images. Unlike text data, image is not searchable without metadata, and to go beyond the most basic analysis of images, it may be best to incorporate the text content. If the text content, even just the headlines of articles, can be overlayed to the image of the newspapers, it would allow a richer analysis of text and non-text elements at the same time. Another way forward is to pursue a spatial analysis of the newspaper, meaning the measurement of the area for a type of content. Such an analysis, properly adjusted to the weight of the page (a front-page article is likely to be more significant than a later page article), would help understand the relative significance of an issue or a person based on the amount of space occupied by it. Still, these comments do not mean that the media montage method is entirely ineffective: the technique was not employed as the singular or final solution to a research problem, but to provide an alternative way of viewing a cryptic and complicated archive. The main method of viewing any given text differently is to read it backwards. Digital tools provide many alternatives, with varying degrees of usefulness.