Future perfect and the ‘will’ you probably won’t have taught

Recently, I was doing some corpus digging to see how the future perfect tense (will have done) is actually used. It’s one of those tenses that tends to make people roll their eyes and question whether it’s worth teaching. And I have to admit, when you see a whole load of examples together on the page of an ELT coursebook, it does start to look awkward and contrived. That’s why I turned to the corpus, to see if I could find some more natural-sounding contexts for it.

I constructed a CQL search as below to look for examples. I knew it wouldn’t capture absolutely everything (it doesn’t include negatives with won’t or passives – will have been done - and it only allows for one adverb in one position – e.g. will probably have done), but it was enough to get me started.

 

Then I disappeared down a grammatical rabbit hole! My main finding? Only roughly 40% of the corpus cites for this structure refer to future time and can really be described as future perfect tenses. So, what were the rest? 

 

Showing my workings:

Before I get into the details, a note about my “methodology”. This was a very quick, rough and ready bit of research, so you shouldn’t pin too much significance on the exact stats. I do, though, think it says something about the general pattern of usage. I looked at three different corpora – the English Trends web corpus (2014-now), the enTenTen21 corpus and for a comparison with spoken language, the BNC2104 spoken corpus (all via SketchEngine). For each one, I performed the same CQL search, then selected a random sample of 100 lines. I went through those lines manually noting down how many lines referred clearly to future time and could be described as the future perfect tense, then everything else was relegated to the “other” category. I did it as a fairly quick scan and I didn’t spend ages agonizing over each line. There were inevitably tricky cases, and I may not have been entirely consistent about how I categorized them. The spoken data was especially tricky, as it always is, because it’s messy and contains lots of repetitions, false starts and fragments of language. This is quite a purposeful construction though (you have to stop and think to put it together), so most of the cites were fairly clear.  There were also a handful  of odd cases in each set where the past participle was actually being used as an adjective with have as the main verb (The supermarket will have frozen fruit, won’t they?), these got lumped in with the “not future”

How DO we use the future perfect?

Looking at the cites which did express a future perfect, they were actually a lot like the kind of examples you find in ELT books. 

 

There were the predictions about the future, especially around climate change (...warn that by as early as 2025 the ice will have melted), or on a more mundane level, weather forecasts (On Thursday that area of rain will have moved away).

There were planning and logistics contexts describing events on a timeline (At 7:30pm, the polls will have closed in Ohio), again, some of them fairly informal and everyday (we're going away the May half term week hopefully - oh okay they'll have finished their exams).

And there were quite a lot of stats and trends of various kinds (... insisting spending will have fallen by £100m over the life of this parliament; The figure will have climbed to 120 by the time the Newcastle outlet opens.)

One interesting usage is in adverts, especially for training courses, touting what you’ll have achieved by the end of the course (By the end of the workshop, you will have developed a draft of a capability statement for your business! When the day is over, you will have experienced Chattanooga in a way that few people have).

I think the main reason that future perfect examples in ELT materials tend to come across as contrived is simply because we don’t typically use lots of instances together. It’s a tense that crops up now and again, so it’s not the individual examples that are inauthentic, it’s just seeing so many of them crowbarred together, especially when someone tries to combine them in a single text … which is hard to avoid when the structure’s the focus of a lesson or activity.

Hypothetical futures

There were a chunk of examples in conditionals or other types of hypothetical contexts, which I don’t want to get into too much here. Some of these were based in reality and fairly easy to pin down as being about future time (If you leave it till tomorrow, they’ll have sold out), others were a bit more abstract and my brain started to melt a bit when I tried to figure out the timelines (if I were sat in the back seat I mean what am I going to do if I think you're driving dangerously? It'll be too late you'll have crashed).

Will as a modal expressing certainty + present perfect

As I said at the start, around 60% of the cites in the written corpora, far more in the spoken data, weren’t referring to future time at all and couldn’t (and shouldn’t!) be classified as future perfect. Taking into account the messy odds and sods, let’s say around half were actually examples of will being used as a modal verb to express certainty, strong likelihood or expectation*, along with a present perfect tense used in the regular kind of present perfect way to express recent past or events at some point in the past with a result in the present. It’s particularly common to make assumptions about what your audience is likely to know already; you’ll have heard/seen/read. But it crops up in a range of contexts, including to express expectations in job adverts (You will have worked at a senior level on at least one national publication; applicants ideally will have completed coursework in technology, software development, transportation or other related areas.)

And of course, will can be used in this way with a range of tenses. Rather than reinvent the wheel, here’s the entry about this usage from the Collins COBUILD English Grammar:

 

 

(Click to enlarge)

 

It’s something I feel I may have seen mentioned in ELT materials here and there, but it’s certainly not a use I’ve seen really highlighted or taught with a variety of verb tenses.  Perhaps one to be added to syllabuses, especially at upper levels?

 

* Of course, there’s an argument that will is always a modal expressing certainty/likelihood. You can replace it with other modals to express different degrees of certainty – By 2030 emissions will/might/may/are highly likely to have exceeded … But at least in the future perfect as described above, it is also expressing future time.