Reducing ambiguity

If you’ve taken an introduction to linguistics, you’ll have heard of ambiguity.  When I say “a cow attacked a farmer with an axe”, there’s some potential for miscommunication about who’s wielding the axe.  The reason is that sentences aren’t just words on a string.  Instead, the words come in clumps, and these clumps can link up in different ways even if the word order stays the same.  It matters if “with an axe” attaches to “attacked” or “farmer”.  This is syntactic ambiguity, but you can also have ambiguity in the words themselves.  If I said “I bought a pen”, you would probably imagine that I’d bought something to write with, but my sentence might actually be about an enclosure for some pigs.  You never know.

In your academic writing, you probably won’t be communicating facts about pigs in pens or cows with axes, but that doesn’t mean that you don’t need to worry about ambiguity.  Here are a few examples that can cause real trouble in essays:

I will present her test of the main prediction of this account, which demonstrates that X has failed to account for Y.

What exactly is it that demonstrates that X has failed to account for Y?  Is it the test?  The prediction?   The account?

Bill notes that Sally criticises Pat.  This is unsurprising.

Is it the criticism that’s unsurprising?  Or is it the idea that Bill took note of the criticism?

This sort of ambiguity can creep into anyone’s writing.  It’s particularly hard to detect when it’s in your own prose, because your brain will choose your intended reading automatically.  So how can you detect problems and fix them?

It might seem a bit mechanical, but running a quick search on your essay can be illuminating.  Doing this won’t find all the problems, and it will list many sentences that are perfectly fine, but I still find the process to be helpful.  Supplementary relative clauses like the one in the first example can be found by searching for [, which], and you can also search for sentences beginning with [This] or [That] if you make sure to turn on sensitivity to capitalisation.   You’ll have to read each instance yourself to judge ambiguity, but it’s a step in the right direction.

Now what can you do to resolve ambiguity once you find it?  Let’s take a look at our sentences again.

I will present her test of the main prediction of this account, which demonstrates that X has failed to account for Y.

A supplementary relative clause coming at the end of a sentence can cause massive ambiguity.  See Huddleston and Pullum 2002 ch 12 sec 4 if you want details about what these clauses look like, but for now I’ll just point out that you could change this relative into a supplementary noun phrase containing a relative (Huddleston and Pullum 2002 ch 15 sec 5 for details).  This sounds tricky, but it essentially means that you could make the thing the relative clause attaches to more explicit.  Depending on what we want to convey, we could change the material after the comma to:

  1. “a test that demonstrates that X has failed to account for Y”
  2. “a prediction that demonstrates that X has failed to account for Y” or
  3. “an account that demonstrates that X has failed to account for Y”

Bill notes that Sally criticises Pat.  This is unsurprising.

The “this” in “this is unsurprising” is what’s called a fused-head noun phrase. Again, see Huddleston and Pullum 2002 ch 5 sec 9 for a full explanation if you’re interested, but it will be enough for now to note that you could put a word after ‘this’ to disambiguate what it is that you’re talking about.  These replacements of the final sentence both make the intended meaning more obvious:

  1. This observation is unsurprising.
  2. This attack is unsurprising.

Joseph Williams has written at length about improving ambiguous sentences like these in at least some of his various books on Style.