Truth, it's generally recognized, is a good thing, and the more true beliefs we have the better. This is because true beliefs tend to work a whole lot better than false beliefs. We invest a lot of time, energy and money in getting the truth. We attend college and university, we subscribe to news sites, we pay consultants, we seek out experts. The more true beliefs we have the more things go according to plan, the more our actions have their desired outcome, and the more we have control over our lives. Perhaps this is why it's said that ‘the truth will set you free’.
One question that's tormented philosophers for centuries is whether we can have true beliefs about the future. It's not hard to see why this would cause headaches. On the one hand, many people have the intuition that a belief – or a claim – can't be true if there's nothing for it to be true about. (For the sake of simplicity I'll use ‘belief’ and ‘claim’ interchangeably here.) Consider the claim that I'll have a salad for lunch tomorrow. When I express this belief today the salad does not yet exist (perhaps the ingredients are in my fridge) and we can imagine lots of things spoiling tomorrow's lunch plans. More generally, the intuition is this: since the future does not yet exist, beliefs about the future cannot be true. (They can't be false, either.) Instead, a belief about the future only becomes true (or false) once the event it predicts does (or does not) actually happen.
But, on the other hand, many people – sometimes the same people! – also have the intuition that beliefs about the future can be true or false, and indeed we depend on these beliefs being true or false. For example, imagine that we ask an engineer if a bridge will support the weight of our truck. The engineer assures us that it will. After we drive safely across the bridge it seems correct to say that the engineer's belief was true and, in fact, it's because it was true that we were justified in driving across.
Looking back, one of the earliest and most famous descriptions of the problem comes from Aristotle. Aristotle’s example was ‘There will be a sea battle tomorrow’ and he argued that, even though it is necessarily true that there either will or will not be a sea battle tomorrow, it does not follow from this that ‘There will be a sea battle tomorrow’ by itself is true or false today. He seems partly motivated by a desire to avoid fatalism: if it is true today that I will eat a salad tomorrow then it seems I don’t actually have a choice when I open my refrigerator tomorrow. Later philosophers raised similar concerns: for example, if God is all-knowing, then He knows what will happen in the future. But if God knows what will happen in the future then there are truths today about what will happen tomorrow and we aren’t as free as we thought (or as free as many religions assure us we are). In order to preserve human free will, as a result, it seems that we must treat claims about the future as neither true nor false.
Claims about the future can quickly lead us into some pretty deep thickets. A seemingly innocuous claim about tomorrow’s lunch can raise thorny theological, metaphysical and logical questions. But here I’ll focus on just one of these questions, namely, whether and how these claims can be true or false, and leave the other questions for another day.
Given people's competing intuitions it's not clear how we should approach claims about the future. Here I'll argue that this question – the problem of ‘future contingents’ – should encourage us to think more broadly about the specific conversational contexts where we make claims and express our beliefs. I'll try to explain why people may have the intuitions they do and why claims about the future aren't as mysterious as they might at first seem.
To begin with it's worth stressing that the problem doesn't arise for all claims about the future but only for future contingents. Future contingents are claims about the future that could turn out to be false: in other words, they are not necessarily true. In contrast, some claims about the future cannot be false: they are necessarily true. For example, it is necessarily true that tomorrow 2 and 2 will equal 4. It’s also true that, if today is Monday, then tomorrow will be Tuesday. Despite being about the future, these necessary truths don’t cause the same sorts of headaches as contingent truths.
Second, the problem doesn't arise for all contingent claims, either. We accept that many claims are either true or false even though we can't know which. Sometimes this happens when the truth is inaccessible, too distant in space or time (or both). For example, assume I claim to have had a burger for lunch on 7 May 2003. In all honesty I don’t know what I had for lunch that day and there’s probably no way of finding out (I don’t keep a food diary). Similarly, I believe that my college-age son is right now studying for his exams, but he's 200 miles away and not answering his phone. In both cases my claim (or belief) is either true or false, but I can't know which it is. There's a truth, but not one I can access because of the distance in time or space. I can still hope that ‘the truth will out’: perhaps I’ll run into an old friend who vividly remembers our lunch from decades ago; perhaps my son will text me photos of what he was doing when he couldn’t take my call. Until then, it seems that my belief that I had a burger for lunch on 7 May 2003, or that my son is right now studying for exams, must be either true or false, though I don’t know which one it is. These beliefs are, in other words, ‘truth-apt’: they could be true. (Not all sentences are truth-apt. Questions like ‘where's my salad?’ and imperatives like ‘close that door’ are neither true nor false; the concept of truth just doesn't apply to them.) Of course some truths are too trivial, or too distant, or too much of a hassle ever to dig up. This is disappointing but hardly surprising. We're not omniscient and human life has its limits. Still, there’s a difference between a claim being true and our knowing that it is true: just because we don’t know that a claim is true doesn’t mean that it isn’t true. Knowability is different from truth.
This means that, with regard to future contingent claims, the problem is not that they are claims about the future or that they are contingent. (There's nothing inherently suspect about either sort of claim.) Somehow, it seems to be the combination of their being both future and contingent that makes them problematic, leading some to suspect that, like questions and imperatives, they can't be truth-apt.
Difference of Kind or Difference of Degree?
Many people (including Aristotle!) have the intuition that contingent claims about the future cannot be either true or false. Of course, intuitions are just intuitions: they can be mistaken, or weak, or flickering. They aren’t the last word, but they can be a useful first word. Consider some intuitive reactions to these claims:
Right Now: I am in my classroom right now.
Near-Term: I will still be in my classroom 5 seconds from now.
Long-Term: I will be in my classroom next July.
On the one hand, I suspect nearly everyone will agree that Right Now must be either true or false. Either the speaker is in their classroom or not (though we may not know which). On the other hand, some people have the intuition that Long-Term is neither true nor false (it is not ‘truth-apt’ in other words): rather, the thought is that it will become true or false next July when the speaker either does or does not enter their classroom. The question, then, is how to think about Near-Term. Is it more like Right Now (true or false; we may not know which) or is it more like Long-Term (neither true nor false)? What do our intuitions say?
It's tempting to treat Near-Term like Long-Term because both are about the future. The thought might be this: there is something that makes claims about the future not truth-apt, something that makes claims about the future different in kind from claims about the past and present. But what could that be? As we saw earlier, some claims about the future are clearly true (‘if today is Monday then tomorrow will be Tuesday’) so merely being about the future is not enough by itself. Or maybe it's this: because the future does not exist, contingent claims about the future cannot be true (or false). But that's not right, either: after all, the past also does not exist yet we generally treat claims about the past as truth-apt. As a result, it's not immediately clear how claims about the future are different in kind from claims about the past and present. In particular, it's not clear why different rules would apply to all claims about the future.
Instead, consider whether claims about the future differ only in degree from claims about the past and present. After all, Near-Term is much closer in time to Right Now than it is to Long-Term. This proximity may be a reason to treat them alike. To see this we can make the differences of degree more extreme:
Very Near-Term: I will still be in my classroom 5 milliseconds from now.
Very Long-Term: I will be in my classroom 10 years from now.
Because the difference between Right Now and Very Near-Term is so small there's no obvious reason to treat them differently: this means that if Right Now is true, then Very Near-Term is very likely to be true as well. (Very Near-Term isn't necessarily true because something could happen in the meantime – perhaps an asteroid strike.) In addition, Right Now provides a reason to accept Very Near-Term: if it's true that I'm in my classroom right now, then that's evidence for it being true that I'll be in my classroom in 5 seconds or 5 milliseconds. In contrast, the difference between Right Now and Very Near-Term, on the one hand, and Very Long-Term, on the other, is great enough to treat them differently. If it's true that I'm in my classroom right now, then that's no reason to think that I'll be in my classroom in a few months or years. The more general point is this: since the future comes in degrees – some events are more in the future than others – it makes sense that claims about the future would also come in degrees. And that would suggest that we can treat some claims about the future – those that are closest in degree to the present – the same as claims about the past and present.
Some might want to go even further and argue that even Long- and Very Long-Term are truth-apt. I'm not opposed to that: I'll argue below that whether a claim about the future is truth-apt depends on features of the conversational context, so it's possible that, depending on the context, some claims about the distant future could be truth-apt. But nothing depends on that here. My goal is to focus instead on the intuition that no future contingent claims are truth-apt. I think that's a flawed intuition, and I want to explain why it's flawed, but I'm not committed to all future contingent claims being truth-apt.
Conversational Features of Future Contingent Claims
If some claims about the future are very much like claims about the past and present then we shouldn't be looking for something in general that sets them apart. Rather, we should be looking for specific factors that might, in some situations, justify treating claims about the future differently. In practice these factors are often features of the specific conversational context where these claims are made: that is, they are the often-implicit presuppositions that help us understand what a speaker is trying to convey. These are the things that, in real-life situations, often go without saying. When we look at claims in the abstract, as we often do when doing philosophy, these contextual features can easily get overlooked. My goal, here, is to make the implicit more explicit.
I can't give an exhaustive list of these contextual features but for present purposes two stand out: (1) the proximity of the future event, and (2) the speaker of the claim. In general, making these features explicit and concrete makes a claim more specific – and the more specific a claim is the more truth-apt it becomes. Again, in everyday situations it's usually clear what sort of time frame we’re dealing with and who is speaking. It's when we consider claims in the abstract that we either disregard these features or fill them in without realizing what we're doing. Let's look at them more closely.
Proximity We've already seen how proximity plays a role in how we judge claims about the future. In general, the more a claim projects into the future, or the longer a time span it covers, the less precise it becomes and the more it is open to different understandings and interpretations. When that happens it’s also more likely to be judged as not truth-apt. As a result we're more likely to view a near-term claim as either true or false, especially when there's solid evidence in its support.
Sometimes the proximity is explicit, as in our example above (‘I will still be in my classroom 5 seconds from now’). Sometimes, however, the time frame is either implicit or unclear. Consider this claim:
This bridge will support the weight of your truck.
If this is said as you're driving up to the bridge and about to cross, then it's natural to treat it implicitly as a near-term claim. But it depends on the situation. Imagine instead that you're working with an engineer designing a bridge on your property and they're pointing to the blueprint. In that case, the time frame is different: they're making an open-ended claim about the bridge's strength once built. Even then it's not an entirely open-ended claim: bridges have life-spans, after all, so the engineer's claim doesn't mean that the bridge will support the weight of a truck in a thousand years (or if you trade in your pick-up for an 18-wheeler).
In these cases, where the time frame isn't explicit, it may not be clear what the speaker is claiming. (‘You said this bridge would last 50 years!’ ‘I said no such thing!’) It's also in these cases, I suspect, that we're tempted to treat a claim as neither true nor false. This is because, taken in the abstract and outside any particular conversational context, ‘This bridge will support the weight of your truck’ is missing important details we need in order to understand what it's saying. Without those details, and without knowing what the claim exactly means, it's sensible to treat it as neither true nor false – in other words, as not truth-apt.
In real life, of course, we can ask for clarification (‘Wait, are you saying this bridge will last 50 years?’), which can make the time frame explicit and may be enough to make the claim truth-apt. We're also adept at filling in gaps by inferring the implied time frame. When someone tells us ‘I'll be in my office tomorrow’ we know that doesn't mean they'll be there for 24 hours; it also doesn't mean that they'll be there for 30 seconds. Between those extremes, however, lie a range of possible time frames and possible interpretations. Perhaps this claim means the speaker will be there during business hours, from 9.00 to 5.00; perhaps it means they'll be there for a few hours in the afternoon. In many cases little depends on the correct interpretation – maybe all that matters is being able to drop something off for the speaker – and so there's no need to make one's understanding explicit. Likewise one's understanding may seem so natural that one isn't even aware of alternatives or that there was even a gap to fill: in that case one's understanding may remain implicit, unspoken and even unthought. Still, different understandings may lead to confusion or misunderstanding, especially when they seem obvious (‘You said you’d be in your office today!’ ‘I was! I was there all morning!’). Clarification helps but we often assume things are clearer than they are, whether in real-life or in philosophy.
Speaker Another factor in judging claims about the future is who, exactly, is making the claim. For example, we're more inclined to treat a claim as truth-apt if the speaker can exercise control over future events. Take a version of Aristotle's example (with the proximity marker ‘tomorrow’ removed):
There will be a sea battle.
Again, in the abstract, it's difficult to know exactly what this sentence is predicting and, as a result, it's hard to say if it's truth-apt. But once we supply a conversational context, including the speaker, it is possible to say more:
‘There will be a sea battle tomorrow’, said the Admiral who gave the order to attack.
In this case we're more inclined to say that the claim is truth-apt. This is because the speaker is in a good position to know that there will be a sea battle tomorrow, and they are in a good position because they have control over whether or not there will be one. (It doesn't hurt that this is also a near-term claim.)
Again, in real-life situations we're adept at filling in features of the conversational context. After all, it's generally clear who is making the claim and in cases where it isn't clear we can ask for clarification. (‘Wait, is that what you think?’) The sorts of examples that give philosophers headaches (‘there will be a sea battle tomorrow’) lack this conversational context and hence leave us adrift, not sure what the claim really means and hence leading us to treat it as not truth-apt. Or else we supply a conversational context (often without realizing that we're doing this) so we can make sense of the claim – but then there's no guarantee that we all have the same context in mind. I might imagine the claim spoken by an admiral while you might imagine it spoken by a fortune-teller and as a result we might disagree over whether the claim is truth-apt.
Control comes in degrees and people rarely have complete control over their future plans. But people do have some control, especially in the near-term. While some might hesitate to say ‘I will have a salad tomorrow’ is truth-apt, these hesitations diminish once we learn the speaker is a habitual and dedicated salad-maker, the sort of person whose meal planning is intentional and not merely aspirational. Sentences like ‘there will be a sea-battle tomorrow’ and ‘I will have a salad tomorrow’ can cause philosophical headaches when, in the abstract, it's not clear who is speaking or what control they have over the situation. Once these features of the conversational context come into focus – once we have a clearer understanding of what the sentence means coming from the speaker's mouth – it's easier to treat future contingents as truth-apt. (‘I will have a salad tomorrow.’ ‘Yes, that's true: you always bring in a salad for lunch.’) Here, too, clarification helps – both in real-life and in philosophy.
Conclusion
The problem of future contingents has been around for centuries, inspiring a range of different reactions and intuitions. My suggestion is that it's less of a problem than it first appears. Once we recognize the difficulty of finding something intrinsic to claims about the future, something that makes claims about the future different in kind from claims about the past and present, we can appreciate some of the ways in which they differ only in degree. In particular we can appreciate how, like other claims, they can be true or false without our always knowing which they are. (Knowability, again, is not the same as truth.) The temptation to treat them as not truth-apt can be explained, at least in part, by the tendency to view these claims in the abstract. Taken out of their conversational context, without the sorts of cues we normally rely on, the meaning of these claims is unclear and, as a result, it's reasonable to treat them as neither true nor false. However, the more we clarify these claims by specifying important contextual features – such as their proximity and speaker – the more they become straightforwardly truth-apt.
Claims about the future are a fun philosophical puzzle but they also raise serious practical issues. We started by noting that truth is a good and desirable thing and that our actions are based on what we believe will happen in the future. If it turns out that these beliefs can't be true then that diminishes their worth and devalues the work we put into acquiring them. It discourages us from taking control over our decisions; it make us less free. Fortunately, that's not an outcome we must accept. By taking the time to clarify these claims we see how future contingents can be true or false (‘truth-apt’) and why it's worth finding out which they are.
Open access
