In Section 1.3, we defined linguistics as the scientific study of language. In this section, we will learn why we consider linguistics to be a science.

Often, when people think of science, they think of the physical sciences, which is the study of the physical properties of our universe. Although some subfields of linguistics, such as phonetics, include the study of physical properties, linguistics generally belongs to the category of social sciences. In the social sciences, we study the properties of societies. We can also categorize linguistics as a cognitive science, which is the study of the mind. Linguistics is the branch of cognitive science that studies what language can tell us about how the human mind works.

The scientific method

All of the types of science — physical, social, and cognitive sciences — are united by their methodology, which is known as the scientific method. The scientific method consists of four parts:

1. Observe data.
2. Formulate a hypothesis.
3. Test the hypothesis against more data.
4. Revise the hypothesis.

The scientific method can be treated as a cycle. After you revise your hypothesis in Step 4, you continue to collect more data, and test and revise your hypothesis, getting more and more precise and accurate with each revision. The second time through the cycle, and every subsequent time, you observe the data you collected from testing your previous hypothesis to formulate a new, revised hypothesis.

An example of the scientific method in morphosyntax

Let’s look at an example as applied to morphosyntax.

Step 1: Observe data

The first step is to observe some data. We will start with the data in (1).

Sentence (1a) includes the pronoun we, which means something like ‘a group of people including the speaker.’ Sentence (1b) includes the pronoun us, which also means something like ‘a group of people including the speaker.’ So an observation we could make about this data is that we and us are very similar in meaning.

Step 2: Formulate a hypothesis

The next step is to make a hypothesis. Since we observed that we and us are very similar, one possible hypothesis could be that they are perfect synonyms.

This hypothesis makes a prediction. If we and us are synonyms, then we might expect them to be interchangeable. That is, we might expect us to be able to switch one for the other without changing the meaning of the sentence.

Step 3: Test the hypothesis

So let’s test out the predictions of our hypothesis! We will take the examples from (1), but switch we for us and vice versa and see what happens.

When we switch us and we, the result, shown in (2), doesn’t sound quite right, so we mark them with an asterisk symbol (*) to show that the sentences are not well-formed.

What have we figured out? Contrary to our first hypothesis, it doesn’t seem like we and us are exactly the same. They are different in some way, which is why switching one for the other in (2) results in ill-formed sentences.

Step 4: Revise the hypothesis

The difference between the sentences in (1) and those in (2) is not a difference in meaning, but a difference in grammaticality. Perhaps, then, the difference between we and us is not a difference in meaning, but a difference in grammatical function.

This hypothesis also makes a prediction. According to this hypothesis, if we change the structure of the sentence but keep the same meaning, it might affect whether we need to use we or us.

Step 5: Test the revised hypothesis

One way to change the structure of these sentences is to put them in passive voice. If we take the sentences in (1) and change them to the passive voice, keeping the same pronoun, we get the sentences shown in (3). If, on the other hand, we make them passive but switch the pronouns, we get the sentences in (4).

This time, the predictions we made in Hypothesis 2 are consistent with our data! Sentences (1a) and (4a) mean the same thing but have different structure, but (1a) must have we and cannot have us, but sentence (4a) must have us and cannot have we. Likewise, sentences (1b) and (4b) also mean the same thing but have different structures and also use different pronouns.

And so on…

Even though our Hypothesis 2 ended up passing our test, that doesn’t mean we’re done. There are two main ways we might continue to refine our hypothesis.

First, notice that our Hypothesis 2 is rather vague. We have proposed that there is a difference in structure, but what exactly is this difference? If we were to continue on with this example, we might try to figure out how to describe the structural difference between us and we.

Second, notice that we’ve only looked at eight sentences. You could even say we’ve only looked at two sentences and variations on those two sentences! That’s not very much data! Another way to continue revising and refining our hypothesis would be to test it against even more data, and different kinds of data, to see if our hypothesis continues to stand.

Check yourself!