The Teaching Tapestry TEFL

Any TEFL teacher with some experience teaching higher levels will be familiar with this scene. You are teaching a classic reading-guided discovery-practice class, and the lesson is going according to plan. The target grammar for today is a thorny topic: reported speech. When it is time to check students’ comprehension of the guided discovery elicitation questions, puzzled and perplexed expressions start appearing on their faces. Questions are being asked, one after the other, and the class is derailed by trying to explain the grammar point. This has happened to me several times, prompting me to wonder why certain grammatical structures are so difficult for students to grasp. Ellis (2008) highlights how, among other factors, L1 interference and complexity of the target grammar likely play a role in students’ understanding of the grammar point itself. As a bilingual speaker of English and Italian, I started reviewing the language analysis and the teaching materials on reported speech, looking for notable differences from my first language that might make it difficult for speakers of Romance languages to understand. That is when I finally noticed something: when explaining reported speech, the word ‘that’ is often omitted. This would never happen in reported speech in Italian or Spanish.

To understand better why this happens and why this is relevant, we need to look at what role the word “that” plays in these contexts.

In English, the word “that” has a number of different uses. One of these is what is called a declarative complementizer, which has the role of turning what follows into a complement. For example:

(a) He told me that he sold his watch.

In the sentence above, “he sold his watch” is effectively the object (complement) of the verb “told”. Compare this to the sentence:

(b) He told me a lie.  

The role of “he sold his watch” in (a) is effectively the same as the role that “a lie” has in (b), which is the object. The word that, in this case, serves the function of declarative complementizer, i.e. it indicates that the embedded clause “he sold his watch” has the function of object. However, as you might already have noticed, if you look at (a), while correct, it’s not what most speakers of English would say as their first choice. Most likely, if you were to actually use (a) in your everyday life, you’d probably say something like:

(c) He told me he sold his watch.

This phenomenon is what we call declarative complementizer omission, and it is by no means limited to reported speech. Indeed, it can happen across a range of different grammatical structures and patterns which will be quite familiar to teachers working with higher level learners. To name a few:

• Patterns of the verb wish: I wish [that] you wouldn’t do that.
• Cognitive and perception verb patterns: I think [that] she is right.
• Clauses where “it” is the subject, often expressing degrees of likelihood: it’s unlikely [that] he did that.
• Other verb patterns, e.g. with the verb “recommend”: They recommend [that] we do this.

Example sentences like the above are common in grammatical explanations of these structures, and the omitted “that” is often not mentioned at all. The reason for this is quite straightforward: general English TEFL textbooks rightfully aim to teach natural language, and they often focus on meaningful chunks of language rather than on more abstract grammatical examples. In addition, writers of textbooks are mostly L1 users of English, and to an English-speaking brain the presence of the declarative complementizer is an optional rather than a necessity in these cases.

Why is this relevant to students’ understanding of these grammatical forms? Well, as I hinted to previously, this phenomenon, while not a rare occurrence in languages across the world, is far from universal. Indeed, in many languages with large pools of learners of English, such as Italian, French, Spanish, or Japanese, declarative complementizer omission is either impossible or much more limited than it is in English. Take the following sentences from Italian, Spanish, and Japanese which are direct translations of (c) (“He told me he sold his watch.”):

(i) Mi ha detto che ha venduto il suo orologio.

(s) Me dijo que vendió su reloj.

(j) Kare no tokei utta to itta.

In all three cases, the declarative complementizer “che” / “que” / “to” cannot be omitted or the sentence would not be grammatical. This is a major difference that is more than likely to create issues in understanding and using target grammar presenting this phenomenon. As a bilingual speaker of Italian and English myself, I can see how, especially at the production stage, justapoxing the two verb phrases “he told me” and “he sold” can lead to misunderstandings and add another layer of complexity to an already complex grammatical structure such as reported speech.

The way to approach this problem is as straightforward as it might seem: explicitly highlight the missing “that” when teaching these structures. Depending on your preferences, you can do this in a few ways. Let’s take the following sentence as an example:

(d) He told me he couldn’t make it.

1. Create a CCQ (Concept Checking Question) to elicit the missing “that”, e.g. what word is being omitted between “me” and “he”?
2. Show the phrase on the board with a missing gap in brackets, e.g.  He told me (________) he couldn’t make it.
3. Simply present the sentence by adding the missing word in brackets, e.g. He told me [that] he couldn’t make it.

Make sure to put “that” in brackets so that there is no confusion over the fact that it’s optional. Grammatically, there is no need to go into too much detail on the role of this “that”, as this will probably only confuse students further. It might be interesting, however, to highlight and have the students reflect on what pragmatic difference there is between using and not using the declarative complementizer: adding that makes the sentence more formal, clearer and deliberate, whereas its omission is more common in colloquial speech. 

It goes without saying that the same exact approach can be used to address complementizer omission in any of the structures where this appears. However, it is also crucial to point out that there are a number of instances where the declarative complementizer cannot be omitted:

• after a number of verbs like reply and shout and, in general, so-called manner‐of‐speaking verbs (whisper, murmur, mutter, grumble, scream, etc.):

(e) She replied that she had already done it. 

(f) He shouted that he was tired.

• Immediately after nouns: 

(g) I am of the opinion that it would be a terrible idea. 

I haven’t had the chance to carry out more formal research on the impact of these strategies yet. However, anecdotal evidence from the classrooms and from colleagues who, after this discussion, started highlighting declarative complementizer omission to students points to a better understanding and use of the target language by learners when this phenomenon is pointed out explicitly. This would also resonate with the fact that choosing to use the declarative complementizer in these sentences makes them clearer. This is because it clearly delineates the boundary between the matrix clause and the “object” (embedded) content, improving clarity. 

If you have been that teacher in that class where students’ puzzled faces were staring at you in eternal perplexion, this is one more trick up your sleeve that will hopefully improve your lesson plan and delivery. As to preventing unexpected and borderline philosophical students’ questions on grammar, there is nothing this writer or anyone else can do I’m afraid!