I like words. I like their abundance. Their variety. The different nuances that they contain. I like how a word like “melancholy,” for example, has a slightly different flavour than “forlorn,” or how the word “myth” conveys something deeper – more emotional, more enduring, more fluid – than its neutral variant, “story.”

Now, most of my Public History peers would probably say that I don’t just like words – I like polysyllabic words. I’ll be the first to admit that long-ish and rather atypical words have a tendency to come to me, unbidden, and that I have an equal tendency to utter them aloud, without thinking. This penchant for the polysyllabic has sometimes even gotten me into trouble, won me unintended notoriety among my peers. 😉

As someone studying in the field of Public History, I realize that I have to think especially carefully about the words that I use, not only because using the “wrong” word can alienate a general audience but also because choosing the “right” word involves weighing various needs, such as those of the client and of the audience, as well as my own for precise language and “good” history. So, while I might immediately prefer the word “auspicious” over “lucky” or feel that “appropriated” explains a situation more clearly than “took,” I’m learning to pause and reconsider the effects of my instinctive word choices, and I’m learning to negotiate the sometimes conflicting needs and desires that exist at the micro-level of diction.

What I didn’t expect was to have to consider the machine as an audience as well. And yet that is what Dan Cohen’s blog post has drawn to my attention. In “When Machines are the Audience,” Cohen suggests that the age of machine-readable text means that we’ll need to write a little differently – or at least with the awareness that what we write can be more, or less, visible in a digital environment depending on the words that we use. The implication is that because text can be read by machines and mined using keyword searches, it is better to use precise and even unique terms or tags in related documents, so that the writing can be more easily searched, grouped, or retrieved. Cohen mentions, for example, how coming up with a unique string of words to identify related history websites can facilitate the process of narrowing searches to these sites only, so that relevant research information can be found. He also cites the example of a legitimate, history-related email being marked as spam, because of its unfortunate use of certain words that are high on the list of favorites for spammers. [1]

Looking over the words I used in a recent email to a friend, I’ll confess that terms ranging from “usurp” and “misnomer” to “vignette” and “quadratic” (yes, as in the quadratic equation) made their way in. (You are probably feeling some pity for my friend right now.) However, I’m consoled, and slightly amused, by the fact that what stands out as atypical word usage is precisely what spam filters ignore. At least, in this area, my penchant for the polysyllabic – for what’s seen as the atypical – has some redemptive purpose. 🙂

[1] Daniel J. Cohen, “When Machines are the Audience,” Dan Cohen’s Digital Humanities Blog, http://www.dancohen.org/blog/posts/when_machines_are_the_audience.