Seeing Cindy’s post which put “As We May Think” in a tag cloud, I started wondering about other text visualization options and understandings they might drive. ManyEyes was long my default for this type of thing but the hassles with Java security have driven me away. So I decided to give Voyant a try. Will Berry1 had used it so well with students, it seemed worth a more in depth exploration.
You also have the ability to select certain words from the corpus2 and they will be charted in relative or raw distribution rates across the corpus. Incidents of “as”, “we”, and “may” are depicted below.
I think the contextual piece is nice, not quite as nice as the branch stuff ManyEyes does but nice and space appropriate. It’s interesting to see that in combination with when the words appeared. Bush seems to stick with “record” almost exclusively and then move into “thought” which is encouraging me to re-read the piece and see if that exposes a deliberate move from something I see as more trivial (record) to more complex (thought).
I also looked at “man” vs “human” because of Morgan’s comment on Bush’s use of “man”. Always interesting to see how time impacts language and how contextualizing writing to its time can change how you read it.3 In any case, it seems like “man” and “human” are used in close proximity and in similar amount with the exception of the middle of the work where “man” is used in isolation repeatedly.
I don’t know if I have any answers on why Bush chose man or human but looking for reasons and playing with rationale has been fun though.
1 #thoughtvectors participant bonus score
2 Humans and works as collections of strange things.
3 I’m sure David could hit me with some interesting sociological perspective on this. Maybe more should be stressed about when “As We May Think” was written. Bush does reference the whole war context in the intro but it may not be clear to all readers what war this is. The time period matters quite a bit as this would not be a very interesting essay if it were written in the 1990s.