Optimality Theory: Systematic Decision-Making

Many linguists learn from their work that pro-con lists aren’t enough to output decisions. Although this method is popular, and popularly perceived as a systematic way of decision-making that an engineer might use, the truth is that pro-con lists are not systematic enough for those who truly desire to maximize the use of their systematizing left brains when they make decisions. Here I shall write about how to use the Optimality Theory approach from linguistics to organize a pro-con table into a usable system of inputs and outputs.

Now, say you’re trying to decide which class of three to take when all of them seem to fit into your schedule. We assume you’ve already eliminated the choices that violate the rules that absolutely, non-negotiably must be satisfied (without which we will suffer the cataclysm of the ages). Under that assumption, let’s first make the proverbial list of pros and cons:

Blah, blah, blah.

To make this into an Optimality Theory violation tableau, we must identify the constraints (i.e., the criteria to try to meet) and the possible outputs (a.k.a. candidates). The possible outputs will go in the far left column after the first row, in any order. [Disclaimer: If you neglect a choice and live to regret your error, you can’t sue me.] The constraints go in the top row after the first column, but these must be prioritized. The logic of this system, it must be noted, presupposes that the ranking of the constraints stays constant: domination in this hierarchy is absolute. If you’re really, really too unsure because two constraints keep swapping places in your mind, make two tableaux.

Okay, are we good?

Mark every constraint violation in this chart with an asterisk. If a candidate violates a constraint more than once, use as many asterisks as the number of times the candidate violates the constraint.

Here’s where we get to making a real decision. Start at the left, which is the top-ranked constraint. Are there candidates that violate it and candidates that don’t? Put an exclamation point after the asterisks where the constraint is violated: these we call crucial violations. If all your candidates violate the top-ranked constraint, too bad. Move on. Do the same thing with every column as you move to the right.

If you have, say, a constraint that forbids killer exams (which we call *KILLER.EXAM), and every course has at least one killer exam, you may see that one has two killer exams, while the others have only one. For this column, the one with two killer exams has made a crucial violation if it hasn’t already made one.

The last man (or woman) standing with no crucial violations is your winner.

If you’re confused, feel free to ask in the comments.


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s