There are two main kinds of arguments, empirical arguments and theoretical arguments. Empirical arguments are based primarily on observation of data, while theoretical arguments are based primarily on ideas. In this class, we will be using both. Often, more complex arguments combine both theoretical and empirical arguments.

Empirical arguments

Empirical arguments are based on data. They have three parts: make relevant observations, make a theoretical claim, and explain how your observations are related to your claim. The parts don’t have to be in this order, but they should be in an order that makes sense.

In this chapter, the fact that children’s knowledge of language is rule-based (as discussed in Section 3.3) is an example of an empirical argument. We observed examples of children’s behaviour, such as their response to corrections from their parents, their pluralization of nonce words, and their use of overregularization. Based on our observations, we concluded that children’s language is rule-based rather than memorized.

Making observations

When you make an empirical argument, you need to start by describing the surface properties of your data. You should be answering the question, “What are the non-controversial facts?” These are mostly observations that a non-linguist could make!

Some examples of observations that you might start which are relevant to morphology and syntax include:

• What does the sentence mean?
• Which nouns and pronouns in the sentence refer to the same entity?
• Is the sentence ambiguous? In other words, does it have more than one meaning?
• What is the word order? Which words precede or follow which other words?
• Are other word orders possible?
• Is the sentence grammatical?
• What similarities are there between this word and other words?

There are lots of observations that you could make about any sentence, but you need to stick to the ones that are relevant to your theoretical claims.

Making a theoretical claim

Although an empirical argument is based on some observation, it leads you to make some theoretical claim about the structure of your data. In this part, you should be answering the question, “What conclusions can you draw based on your observations?” This is where you talk about things you can’t see or hear, but that you infer based on what you do see or hear.

Some examples of theoretical claims you might make in this course include:

• Whether a word is simplex or complex.
• The classification of a word or sentence based on its structure.
• The part of speech of a word (e.g., whether a word is a noun or a verb), as we will learn about in Chapter 5.
• Whether a string of words behaves as a unit, which we will learn about in Chapter 16.

Backing up your claim

To make a complete argument, you also need to explain how your your observations provide evidence for your claim. You cannot assume that your readers will be able to draw the same conclusions as you!

In some cases, you should also think about other claims people might think to make about the same data, and why your analysis is better.

Theoretical arguments

Theoretical arguments are based on ideas, logic, and reasoning, instead of data. Because of this, we can never trust them as much as empirical arguments. However, we wouldn’t be able to get as far in science without theoretical arguments, too!

In this chapter, the Poverty of the Stimulus argument in Section 3.4 is an example of a theoretical argument.

A theoretical argument has two parts: premises and a conclusion.

Laying out your premises

A theoretical argument is based on premises, which are the ideas that you use as a foundation to draw your conclusion. They are your claims about how things work. A premise can be a conclusion from a previous (theoretical or empirical) argument, or it can be an easily observable fact (e.g., language is infinite; humans acquire language; not all languages are the same).

For your argument to work, you need to be relatively certain that your audience will accept your premises.

The Poverty of the Stimulus argument has two premises: that language is infinite and that infinite systems are unlearnable.

One of the ways to argue against a theoretical argument is to claim that one (or more) of the premises is false.

Drawing a conclusion

Your premises should logically lead to a conclusion. It’s important that you actually state what that conclusion is, though, instead of making your readers figure it out.

Another way to argue against a theoretical argument is to claim that the conclusion does not actually follow from the assumptions.

Consider your assumptions

Every time you make an argument, you make theoretical assumptions. These can be foundational, like “I am assuming generative grammar is correct,” or structural, like “I’m assuming that two things can’t occupy the same spot,” or methodological, like “I’m assuming that grammaticality judgments are reliable.” A lot of times, we are making assumptions without realizing it, so it’s always good to spend some time thinking about what assumptions we are making.

Decide whether you should be making your assumptions, or whether you should dig deeper and investigate some of them. Sometimes it’s useful to make assumptions, though, to see where a theory will lead us, or because if we had to start from scratch with no assumption every time, we wouldn’t make it very far very fast.

Decide whether you need to state your assumptions. You don’t need to state any assumptions that you are pretty sure your audience shares. This happens a lot in a class setting—we share the assumptions of the class materials. You also don’t need to state any assumptions that are pretty obvious from your argument. But beware! Sometimes what is obvious to you will not be obvious to your reader. When in doubt, state your assumptions.

Presenting your argument

When you write an argument, you should clearly state all of the components of your argument and how they are connected.

Sometimes you will have to make a chain of arguments. For example, if we were to argue that a language has prepositions, we first have to identify an adposition and which noun it goes with, and then show that the adposition goes before that noun.

Sometimes a chain of argumentation will include both empirical arguments and theoretical arguments. For example, our argument that children use rule-based language actually has both empirical and theoretical components.

As you lay out all of the arguments in your line of reasoning and each of their components, think about what order it is best to lay out your claims. Do any of your points depend on another point?

Always make sure you use technical terminology where appropriate. Technical terminology is useful because it is normally concisely defined to avoid ambiguity. For example, saying un- is a morpheme is more precise than saying un- is part of a word.

Check yourself!