Can there be cause and effect without time?

Our usual understanding of cause and effect operates tensely; that is in time. But consider a basic statement in some generic computer language:

if x then y

Is this in time? Well an actually computer operates in time so this statement is also in time; but one can understand the program, in principle, without operating the computer; thus it appears we have some notion of causality without time; I'm not asserting that this is indeed the case; but to motivate the question - can there be causality without time?

One might turn this question around and say that where thee is causality there is time; but it may be a time unlike that which we actually experience depending on the structure of causality being defined.

In fact, it turns out that some general features of spacetime (of GR) can be deduced from simply the causal structure which lends some credence to this idea.

• A photographic example: With time being suspended, it would be difficult (impossible?) to determine cause and effect from a single, still photo alone. It would be much easier with a video as a sequence of photos over time. – Dan Christensen Aug 24 '14 at 14:17
• Possible duplicate of Does causality always require an arrow of time? – Conifold Feb 21 '18 at 3:04
• @conifold: Why are you pointing out that this 'question is a duplicate', when the said question was written in 2016, two years after this one. Aren't you turning causality backwards here? Can you please remove the duplicate tag. – Mozibur Ullah Feb 22 '18 at 6:31
• Because it has more upvotes and an accepted answer. The point of marking duplicates is to organize information and simplify navigation for users, not chronology. – Conifold Feb 22 '18 at 18:44
• @conifold: I tend to think chronology matters too. I wouldn't call it a duplicate but merely a question that touches upon the same topic. – Mozibur Ullah Feb 26 '18 at 9:32

I think that you've implicitly answered your own question: in terms of colloquial, macroscopic causuality, as far as we have observed*, the causal agent needs to precede the effect temporally; If you consider specific formal extensions of the naive concept of causality, in certain physical theories, then you can end up with the idea that time (temporal ordering of events) arises from considerations of the causal structure of the scenario at hand. In any event, causality and time are intimately connected to one another.

I believe that you're computer language example is misleading in its terseness. In an imperative computer language `if x then y` means (more fully) `in the case where x evaluates to non-false, then evaluate statement y`; which clearly involves time; note that `then` is used in one of its temporal/causal senses. This is a problem inherited from English (and other languages?) where it uses the same structure `if x then y` for both causal (and thus temporal) implication and for a strictly logical relationship, which in my interpretation is just a particular truth-table and lives outside of time, except through any temporal features of the two variables.

`*` this caveat is here due to conceptual possibility of time-travel, but even that might be interpreted as the causal agent time-traveling to a point prior to, but arbitrarily close to, the time of the effect.

• Remember also that something must have happened before or X would have no meaning. Even if "before" was "before the program started" or "before the chips were printed". (Or we must be willing to stop and wait as with blocking i/o functions.) An if statement always occurs in a specified or happenstance sequence with other directives. Otherwise it would be "when x, do y", a directive we can only emulate. – StarWeaver Jan 24 '18 at 12:02

I think causality can be thought of independent of time. Indeed, this is necessary when considering time travel, and the paradoxes that arise there: In time travel, you have future causes with past effects, quite the opposite of the normal order. If the concept of cause and effect would be bound to the concept of time order, then we would not even be able to think about time travel. But we obviously are able to think about time travel. Note that this is independent of the question whether time travel is actually possible.

Indeed, the concept of cause and effect is easy to describe without making reference to time at all: If you can determine Y by manipulating X, then Y is causally dependent on X.

• Doesn't time travel involve the notion of time ? All that time travel does, as does the idea of backwards causation, is to drop the assumption that a cause precedes - is temporally prior - to its effect. It reverses the time order but does not abolish it. – Geoffrey Thomas Jan 23 '18 at 11:08
• "Indeed, the concept of cause and effect is easy to describe without making reference to time at all: If you can determine Y by manipulating X, then Y is causally dependent on X." You've made a reference to time inside the word "manipulating". – kralyk Sep 9 '18 at 19:38

In ordinary language, we might say things like "the bath water is draining because there is no stopper in the drain". The state of affair "there is no stopper in the drain", then, causes the state of affair "the bath water is draining". On a states of affairs view of causation, time doesn't need to enter into it.

(Notice if this last example worries you because of the action word "draining", use another example: the air conditioning being on causes the room to be cold.)

It's more standard in the philosophy of causation, though, to say that causation is strictly a relation between events. On that understanding, time will certainly be involved, since it doesn't make sense to talk about a timeless event.

As for the computer question, insofar as causation occurs at all here, it is sequential. The computer determines the truth value of x, and then if it turns out true, it proceeds (in time) to evaluate y.

This is partly a computer science question. Your if-statement is an instruction to a computer to perform y when condition x is true. Computer programs run very quickly, but they run in time as distinct events, causally related to one another.

Computer scientists though, when they analyze programs, do not treat them as causal structures made up of events but as logical structures made up of propositions. Your if-statement, in this context, becomes a statement not of what happens but of what must be true.

Computer scientists use this kind of analysis to prove that a program does what it is supposed to do. This proof is a mathematical or logical proof. It relies not on causal necessity but on logical necessity. Propositions are not in time in the same way that events are in time. For example a syllogism, if it is ever valid, is always valid. Its validity is not momentary or ephemeral the way causes are.

You ask "is this in time?" The answer is--it depends whether you are treating the statement as an instruction that is performed or as a logical proposition. If it is a performed instruction, it is in time, but if it is a logical proposition, then it is not causal.

I suppose a third way of treating the statement is as a law--whenever x happens, do y. Laws are the principles that govern causes. Are laws in time? I would say that they are also in time. In the case of a program, the law exists only while the computer is running the program, that is, at a definable time.

I do not think you have found an example of causality that occurs without time, and I cannot easily think of one myself. The only thing that suggests itself to me is the sequence of decisions that occur in the mind of an eternal being (where eternity is conceived of as somehow outside of time). Each decision depends causally on previous decisions without being logically necessitated by them and without happening in time. This is an example of what you are looking for, but not one whose existence we can easily verify.

1. 'If X then Y' as you introduce it is a logical relationship without reference to time. I might say, 'If X (this is a triangle) then Y (this has three sides and three internal angles)'. There is no notion here that something's being a triangle causes it to have three sides and three internal angles. The relationship is, if anything, purely definitional. In programming logic, I assume 'If X then Y' is an instruction for the program to do Y in condition X.

2. There are accounts of causation, such as Aristotle's theory of the four causes, which are not strictly or purely temporal : the material cause, the formal cause, the efficient cause are in time but the final cause (to ou heneka, to telos) is not. See e.g. Aristotle, 'Physics', II.3, 194b and 'Metaphysics', II (a), 994b. Also in Plato's 'Phaedo' the Forms, transcendent of space and time, are recognised as causes ('Phaedo', 105b-c).

3. The standard idea of causation is that a cause precedes, is temporally prior, to its effect. But (a) if heating water to 100C causes it to boil, doesn't it boil at the instant (not after) it reaches 100C ? However, this doesn't exclude temporality from causation, it just allows the cause not to precede the effect. Also (b) there is the possibility of backwards causation. This also doesn't exclude temporality; it simply allows the future or present to influence (not necessarily to create de novo) a past event or state of affairs.

4. Conclusion : (i) your opening ''If X then Y' is logical, not causal; (ii) causes can be simultaneous with their effects not necessarily prior to them. (iii) the time order can be reversed in backwards causation. (ii) and (iii) preserve the temporality of causation. But Plato and Aristotle operate with ideas of causation in which a temporal dimension is excluded - Plato's Forms are non-temporal, transcendent of space and time, and Aristotle's four causes includes a non-temporal element in the final cause.

As reference I will use Stephen Mumford & Rani Lill Anjum’s “Causation: A Very Short Introduction”. Although they do not discuss causality without time, their survey provides insight that should be useful in viewing causality atemporally such as through the creative contemplation of Plotinus.

These authors start with Hume’s analysis or reduction of causation into three simpler ideas: (1) repeatability, (2) cause comes before effect, and (3) constant conjunction. Kant criticizes the second, claiming that causation happens simultaneously, however, this still leaves causation within time. Others use necessity or a transfer process or some other method to analyze causation differently, but all of them miss some aspects of what is intuitively meant by cause. The authors provide counterexamples for all of these theories.

More recently people tried to combine the various reductionistic causation theories into a pluralist setting, but that simply emphasized the failure of analysis to reduce causation to something simpler. Finally the authors presented the idea that causation is a primitive concept. It cannot be analyzed into simpler components. The components used by other theories to reduce it to something non-cause-like are seen by the primitivist theory called dispositionalism to be only symptoms that a cause is present. With dispositionalism singular objects have real dispositional properties or causal powers.

Dispositionalism is an old idea going back at least to Aristotle. Hume was reacting to this tradition when he tried to analyze causation without admitting the existence of causal powers. Dispositionalism provides a justification that lets us see causation as a primitive idea. It also points back to ancient philosophy which may provide a link to someone like Plotinus who had the idea of creative contemplation that is active or causal but also atemporal. In this way one might be able to construct an atemporal theory of causation.

I can say "if a wind blows trees sway" and "if trees sway a wind blows". In our "normal" world the first statement deals with cause and effect, the second one can be just a phenomenological observation. Without time the both should be declared as direct causal relations by the structure of the statements. I am not sure that the world thus constructed will be equivalent to our one.