Let's read: Bringing about the past, and Backwards causation

Backward causality is not something only physicists think about. Philosophers have thought about it a lot. We'll read about a classic paper Bringing about the past (1964), Michael Dummett.

I'm going to just pick out memorable and easily understandable tidbits that collage to a picture of what the theory of backward causality is about, and not try to summarize the whole thing -- if you want the whole thing, read the paper in whole.

The mystery of praying for the impossible

The Efficacy of Prayer (1959), C. S. Lewis 

“God,” said Pascal, “instituted prayer in order to lend to His creatures the dignity of causality.” But not only prayer; whenever we act at all He lends us that dignity. It is not really stranger, nor less strange, that my prayers should affect the course of events than that my other actions should do so… 

Here, "dignity of causality" roughly means something like this: to live with dignity, a human need to have the ability to cause things to happen. It's Pascal's response to "why do anything when God can do everything?", replying, "God made humans do a lot of hard work because it gives them dignity", a classical case of theodicy. But... what if the prayer is about things in the past?

Dummett included a discussion on "retrospective prayer", which is the best story in the paper.

... [according to orthodox Jewish theologians] It is blasphemous to pray that something should have happened... it is logically impossible to alter the past, so to utter a retrospective prayer is to mock God by asking Him to perform a logical impossibility. 

Wow, orthodox Jewish theologians are admirably logical about blasphemous prayers, despite how intuitive the idea is.

... suppose I hear on the radio that a ship has gone down in the Atlantic two hours previously, and that there were a few survivors: my son was on that ship, and I at once utter a prayer that he should have been among the survivors, that he should not have drowned; this is the most natural thing in the world... Are the Jewish theologians right in stigmatizing my prayer as blasphemous?

As Dummett drily remarked, the retroactive prayer is the only case where a normal pony can attempt to cause things to happen in the past, without being regarded as an absolute madmare:

.. it is the only instance of behavior... designed to affect the past and coming quite naturally to us. If one does not think of this case, the idea of doing something in order that something else should previously have happened may seem sheer raving insanity. 

Dummett suggested a way to resolve this paradox: if praying for a future event makes sense, then praying for a past event can also make sense. Imagine that Celestia lives in meta-time, and some parts of the past do not yet ("yet" as in meta-time) exist, like the future, then praying is not asking Celestia to change the past (impossible), but simply, when She finally gets around filling in the details of the past later ("later" in meta-time), She would ("would" in meta-time) fill it in so that the past is as I prayed.

Another way to resolve this (which I prefer) is full determinism:

God knows that I am going to pray that my son may not have drowned because He is going to make me pray so... everything that happens is directly effected by God, and that human freedom is therefore confined to wholly interior movements of the will. 

This is the view adopted by Wittgenstein in the Tractatus, and there expressed by the statement, "The world is independent of my will." 

Smol digression: do prayers work?

There have also been scientific of retroactive prayer, with dubious results: Retroactive prayer: lots of history, not much mystery, and no science (2004), Jeffrey P Bishop and Victor J Stenger.

Leibocivi (2001) reported a study that showed prayer done for patients well after they had left the hospital, had reduced the length of stay in hospital and duration of fever from blood stream infections. In short, prayer somehow seemed to act backward in time to shorten patients' stay in the hospital...  

Olshansky and Dossey subsequently argued that a logical explanation might be found for Leibovici's results. They point to numerous other randomised controlled trials to support their thesis that prayer could work at a distance of space and that it might be plausible that prayer could act retroactively in time. We argue that their claim is built on a confusion and lacks a deep physical model... 

and they finally conclude, surprise:

Current scientific theory does not support effectual benefit of prayer distant in space or time.

The story of the dancing chief

Dummett... [imagined] a tribe [the Maasai?] whose men participate in lion hunts to prove their bravery. Hunt observers report to the tribe’s chief whether the men indeed behaved bravely. After the hunt ends but before the observers have reported, the chief performs dances intended to cause the men to behave bravely...

Dummett proposes an experiment to test the chief’s belief in his ability to affect the past. Noting that the chief has never danced after the observers have reported, Dummett has us challenge him to dance only after receiving the report, on an occasion when the men have reportedly not been brave. Dummett describes three possible outcomes of this experiment. 

First, the chief dances after learning of the men’s cowardice. This shakes his belief in the correlation between dancing and bravery. 

Second, he tries to dance but fails. This challenges his belief that his dancing causes the bravery; it suggests, instead, that the bravery makes the dancing possible. 

Third, he dances and later discovers that the observers misreported: the men behaved bravely, after all.

If the first two outcomes are rare, then the chief's belief would remain intact.
Though in any case, they have got to stop hunting the poor endangered lions.

We reject fatalism, and the chief rejects retro-fatalism

Dummett noted that the usual argument against retroactive prayer is very similar to the fatalism argument, he then wrote a lot of words rebutting the fatalism argument the one about not bothering to change future events, since the future events will happen. But!

[during the bombing of London in 1940] The fatalist argues, "Either you are going to be killed by a bomb or you are not going to be. If you are, then any precautions you take will be ineffective. If you are not, all precautions you take are superfluous. Therefore it is pointless to take precautions." ... the argument is formally quite parallel to the argument supposed to show that it is pointless to attempt to affect the past; only the tenses are different.

... We say to the chief, "Why go on dancing now? [You can't change the past, since the past events have happened]" ... The chief can reply in exactly the way in which we replied to the fatalist... [a long argument that is essentially the same as the argument against fatalism, but with past and future switched] ... This reply sounds sophistical; but it cannot be sophistical if our answer to the fatalist was correct, because it is the exact analogue of that answer. 

Rebecca Roache

What is it like to affect the past?, Rebecca Roache: It first summarized Dummett's article,

whilst backwards causation is possible, the possibility of actions being preceded by their intended effects is subject to a condition: if agents are to be capable of affecting the past, then prior to acting they must believe that they cannot know whether the intended effect has occurred. 

So causality, something seemingly fully objective, can somehow depend on subjective beliefs...

Dummett describes a scenario in which this condition is satisfied, and concludes that it is zconceptually possible for actions to affect the past... [but] the attitude of agents who believe themselves capable of affecting the past must seem ‘paradoxical and unnatural to us’.

Then Roache claimed the opposite:

I will show that the price for holding that agents could affect the past is not as high as Dummett supposes. Far from seeming ‘paradoxical and unnatural’, such agents’ adopting the attitude Dummett describes can be shown to be reasonable. 

Agents like the Heptapods are very inhuman, and yet they make their own kind of sense. Sometimes I feel like a heptapod, or do I? Is it too difficult for a human to feel like a heptapod? Roache argued that in some worlds it'd be not difficult to hold such attitudes, and yet still be humanlike, but I didn't read that.

Heptapod psychology

Dummett noted that a rational agent cannot hold three beliefs together: at least one must be given up.

If anyone were to claim, of some type of action A, (i) that experience gave grounds for holding the performance of A as increasing the probability of the previous occurrence of a type of event E; (ii) that experience gave no grounds for regarding A as an action which it was ever not in his power to perform; (iii) that it was ever possible for him to have knowledge, independent of his intention to perform A or not, of whether an event E had occurred.

In less abstract words: If the chief believes that

1. Dancing after a hunt increases the probability that the hunters performed bravely.
2. He can dance if he wants to.
3. It's possible for him to know, after a hunt, but before dancing, if the hunters performed bravely.

Then he must be irrational, because experiments show that these three beliefs cannot all be true.

Any rational agent who believes in their ability to cause things in the past must believe in 1 and 2, thus cannot believe in 3. Dummett found this possible, but very weird and unhumanlike.

The Bilking Argument

Roache noted that the story about the chief is specifically intended to rebuke the 'bilking argument' against backwards causation (Black 1956; Flew 1956, 1957 and 1964).

But what the hay is the bilking argument? Let's read Backward Causation (2015), Stanford Encyclopedia of Philosophy:

It started by noting that the philosophy of retrocausality is quite new, becoming "popular" only in 1954:

The philosophical debate about backward causation is relatively new... No empirical phenomena seem to demand a notion of backward causation for our understanding of them. And for a long time it was thought that such a notion involved either a contradiction in terms or a conceptual impossibility... In 1954 Michael Dummett and Anthony Flew had a discussion about whether an effect can precede its cause. Dummett defended the idea whereas Flew argued that it involved contradictions in terms.

Two years later, Max Black (1956) presented an argument against backward causation, which became known as the bilking argument, and later attempts to meet the argument seemed to generate all kinds of paradoxes. 

The bilking argument is basically the reverse of grandfather paradox: if B happens now, and A happens in the future, and A causes B. Then we can just stop A from happening, and that is a paradox! Hence, we cannot have backward causation.

In story form: If your parents conceived you only because you would in the future travel back in time and kill your mother's fiance, forcing her to marry your father (the second-best choice), then it's imaginable that you would instead decide not to travel back in time at all, creating a paradox.

Newcomb Paradox

And of course, there's the Newcomb paradox. In the world of Newcomb paradox, there's a lot more reason to believe in backwards causation. I may write about this later.