Assigning human characteristics to non-human things is a time-honored writing technique. Most writers are not only aware of and familiar with the idea of personification—describing something as a person when it definitely is not—but at some point in their educations they’ve been encouraged to use it. Countless writers have used the technique before them: Homer (“rosy-fingered dawn”) and Shakespeare (innumerable times, one of the most well-known being “But look, the morn in russet mantle clad/Walks o’er the dew of yon high eastward hill”); but you can hardly pick up a work of fiction and not find examples. Tolkien, Fitzgerald, Roth, Atwood—take your pick of author, genre, or era. You’ll even find it in titles: The Moon is a Harsh Mistress by Heinlein readily comes to mind (and that example is both personification and metaphor—there’s overlap in literary technique).
There’s a similar idea called anthropomorphism (other forms: anthropomorphizing, anthropomorphization), which likewise assigns human characteristics to something not human. In fiction, this technique can come in handy. In non-fiction, especially journalism and science writing, however, not so much. In those cases, it’s long been argued that anthropomorphism is sloppy writing and bad form.
How can you tell you’re anthropomorphizing? And how can you stop doing it? Let’s take a look.
The difference between personification and anthropomorphism can be thought of in some ways as the difference between simile and metaphor: a simile states that something is merely like another thing; metaphor declares that it is the thing. Likewise, anthropomorphism assigns human characteristics to a thing, but personification treats that thing as a person. (This is a simplified and not entirely accurate explanation, but for the purposes of this post it’s adequate.)
A few months back an article crossed my desk which, as an editorial professional, I viewed as important enough to share. “In the animal kingdom, what is promiscuous?” appeared in the September 15th, 2018, issue of Science News and discussed a study (unpublished, but the author’s earlier thoughts on the subject can be found here) which was presented at the Animal Behavior Society’s annual conference. In that study, behavioral neurobiologist Sarah Jane Alger looked at the use of a single word, “promiscuous,” in close to 350 studies of animal mating behavior. Her analysis revealed a few things, some specific to that particular word, some more generally applicable to the concept of anthropomorphism. In particular, Alger showed that using a word like promiscuous is potentially confusing because different researchers use it to describe different behaviors. It is, Alger and others argue, a poorly defined term that leads to imprecise science. Not only is the word used inappropriately, but as work Alger builds on (Elgar, Jones, and McNamara) reveals, the use is often sexist: it’s applied to female animals, with all the baggage the word brings from human behavior, far more than to male animals—despite the fact that in nature “females…remain choosy, irrespective of the number of mating partners.”
That specific word aside, this study highlights a perennial problem in scientific writing and science journalism: anthropomorphizing. Assigning human characteristics to animals (or non-living objects) is always, at best, a risky proposition. A word that describes a particular type of human behavior, such as promiscuous, might be completely appropriate in a poem or in a passage of description in a novel. But it has no business in good scientific writing for two reasons. First is that the previously noted imprecise definition should sound an alarm that a better term should be used. But second because the term, by aligning with a particular human behavior, also suggests that the animal is acting in a human way: use of the term anthropomorphizes the non-human subject.
Elgar et al. express the problem bluntly (highlighting mine):
The discipline does not tolerate other anthropomorphisms in biological science; for example, the term forced copulation is preferred over rape, and infanticide preferred over murder. Promiscuity has pejorative and androcentric connotations and is likely to be emotionally evocative, typically saved for the females of the species: while polygynous males maximise their fitness by mating at the highest rate, females are described as promiscuous.
They conclude, succinctly, that “promiscuity can be replaced with polyandry, polygyny and polygynandry, as appropriate – descriptive terms that are silent about the nature of mating, and devoid of sociological, psychological and moralistic connotations.” Or, as Alger puts it, “It’s bad enough when we judge each other – let’s try not to judge animals too.”
Anthropomorphism is generally frowned upon in good science writing, and rightly so. But it can be difficult to stamp out. It’s very easy for humans, including writers, to see human-like behavior in the actions not only of other animals but also of objects and natural phenomena. It’s easy to carelessly attribute a human-like action where it has no business: to write, for instance, that a virus prefers cool weather, that cichlids want to play, and that gut bacteria wage war.
(Even those articles cover a wide range: that virus is only anthropomorphic in the title; it nonjudgementally “thrives at lower temperatures.” The gut bacteria, however, not only “wage war” but “stake out space,” “attack each other,” and “cooperate,” presumably as their current mood or foreign policy dictates. The fish…well, that summary article gets pretty anthropomorphic, but the abstract of the original paper it summarizes is much more carefully worded.)
Here’s a brief and interesting take not only on how anthropomorphizing organisms mischaracterizes physical reality, but also how it potentially misdirects and stunts scientific inquiry.
The APA frowns on anthropomorphism. APA style notes that “the study found” is an anthropomorphic (and incorrect) way of saying “the researchers found.” The study is an abstract concept, not a living entity, and cannot find anything; the researchers deserve the credit. That sort of construction, however, is so deeply embedded in science writing that it will probably never be stamped out (and I have few doubts that you’ll find examples similar to it on this blog, perhaps even in this very post, despite my best efforts).
I mentioned that I’d suggest a few ways to avoid anthropomorphism in your writing. To reiterate, there’s nothing wrong with using it in fiction, or even in non-fiction when creative license is allowed, but in some places, most especially science writing, you should try to avoid it. How do you do that?
The easiest method is to steer clear of attributing human characteristics to non-human things: don’t talk about objects or other animals as thinking, wanting, deciding, preferring, attacking, and so forth. Simply heeding that advice will take care of the vast majority of potential problems.
After that, check all the verbs (actions) and make sure that there’s a human agent involved. Have you written that “the data prove” or that “the literature recommends” something? Reword those constructions to make it clear that a person or persons are taking the action: “Wilson proved” or “the authors recommend.” Sometimes it’s as simple as that. APA style includes exceptions for three verbs: show, present, indicate. Academic objects—studies, theories, data, results, and the like—can use these verbs. Others, if you’re strictly adhering to APA style, are used at your peril.
One last suggestion for avoiding anthropomorphism is in line with something I suggest frequently: say what you mean to say, clearly and concisely. Modern writers in American English have a tendency to use more (and longer) words than they need to. This problem, in my unscientific observation, has become noticeably worse in recent years. Academic writing follows certain predictable forms, but it doesn’t have to be bad writing. Sadly, bad and over-wordy writing is all-too-often encouraged and rewarded: if it’s not obscure and difficult to understand, it must not be worth understanding. The trend is even worse outside of academia. I view that as a side effect of the times we live in, when education has been devalued and intellectual endeavor is ridiculed more than we’ve seen for at least a couple of generations. In that climate, it’s only to be expected that a default measure of how “smart” someone is isn’t what they actually know, but how ornate and convoluted their presentation is. But that’s another issue: the suggestion for avoiding anthropomorphism in your writing is to be direct. Don’t fly off in over-wordy loops and whorls; instead, keep your ideas clear, your nouns and verbs close to each other, and the ideas and information you want to convey tight and well-contained.
For a very clear description of the problem of anthropomorphism in academic writing and a nice variety of examples (positive and negative), check out this article. Part of the discussion, when it gets into hairsplitting, might give you a headache, but if this is the kind of thing you need to know, then you need to know this kind of thing. It’s worth pointing out that while this is focused on science writing, the basic principle (and at least one of the examples) applies to good business writing as well—even if you’re far less likely to run into trouble with it in that domain.
thebettereditor 