Sunday, March 06, 2016

Break the bank

1. One line of evidence for God's existence involves examples of special providence. This might include modern miracles and answered prayers. Likewise, there are things we will need in the future, but we don't know that in advance. We'd pray for it if we knew we were going to need it. So in some cases God might provide for us as if that were an answer to prayer, because we don't know ahead of time that we need it to happen, and by then it would be too late to pray. 

Now in some cases the windfall might be consistent with special providence or luck. Chances are, you will get lucky every so often. Coincidences happen. But I have in mind examples that are highly resistant to naturalistic explanations. Where it's too specific, unlikely, and opportune to be sheer luck.

2. However, "skeptics" discount this evidence as sample selection bias. The distribution is random. It averages out, when you take everything that happens to you into account. For instance, sometimes you get what you pray for, and sometimes you don't. Some people are healed, and some are not. If you only compare healings, it looks impressive. If you add dissimilar outcomes, it all blends into the undifferentiated background. Or so goes the argument. 

3. There are, however, at least two major problems with the "skeptical" objection. To begin with, it backfires.

Suppose there really is a pattern. If, however, our sample is too small, then there's no reason to expect a discernible the pattern. If all we have to go by are anecdotes and isolated incidents, then it would hardly be surprising if the pattern entirely escapes our notice, for it only emerges if we have a much larger sample. In that case, apparent randomness is perfectly consistent with a deeper, broader pattern. So the very thing the "skeptic" mentions to show it's really random is the same thing that's consonant with its nonrandomness. 

In terms of reported miracles, answered prayers, and other special providences, our provincial knowledge is only skimming the surface. We know next to nothing about what most other Christians experience at different times and different places. So even if there were a pattern, how would we be in any position to perceive it? 

To take a comparison: suppose I'm a Martian who's assigned to study human behavior. I see a family of four load the trunk of their car with luggage and drive away. If their objective is to reach their destination, then they will take the shortest route. Depending on the length of the journey, they will drive as far as they can each day. Their route will be determined by the location of motels, gas stations, and the distance between the starting-point and the end-point. 

Yet my Martian logic is confounded by their actual behavior. They don't travel in anything like a straight line. They constantly veer off. They may stay in a town or campsite for several days before they resume the trip. To all appearances, their behavior is random.

But from a human perspective we know that's probably not the explanation. Rather, this is typical tourist behavior. Their objective was never to simply reach their destination. Rather, it was always more about the journey than the destination. They are sightseers. They drive on scenic routes. They visit historic towns. Far from being random, their trip is meticulously planned. Where they will go. How long they will stay. Each day is accounted for. 

In addition, our Martian can't tell from where they begin what their destination will be. He doesn't know if they plan to drive 50 miles, 500 miles, or from coast to coast. They might head east to west for most of the trip, then turn south during the final leg of the trip. Our Martian observer might have no inkling three-quarters of the way through the trip where their intended destination is. To register the pattern, you need to begin at the end and work backwards. 

And it could be the same way with providence. The pattern defies recognition if all you have are isolated data-points. 

4. However, the "skeptic" might object that this only shows, at best, how the phenomenon is consistent with either randomness or nonrandomness. Mind you, even if that were the case, it greatly attenuates the original objection. According to the original objection, what we really have is evidence of randomness, once you take all the evidence into consideration. But now the "skeptic" must concede that the distribution pattern isn't evidence for randomness–appearances to the contrary notwithstanding. 

5. But it's not just parity. As I noted at the outset, what if you have examples of special providence which are not plausibly susceptible to naturalistic explanations? Then that's positive evidence for special providence. 

To take a comparison, suppose a group of ten Caltech students or MIT students decide to break the bank. They figure out how to cheat casinos. They do it as a test of ingenuity. Perhaps they hack into the security cameras so that they can actually see the poker hands, and they devise some undetectable signaling system.

They divide up into teams of two and hit five casinos in Las Vegas. The same team never goes to more than one casino, so there's nothing to directly connect the group of ten cheaters. 

It doesn't take long for each casino to catch on to the fact that something is afoot. A player is beating the odds way too often for that to be coincidence. Yet these are isolated incidents. 

Suppose each casino is ignorant of the fact that four other casinos are encountering the same thing. Or even if they knew it, they have no background information on the players to connect them. Even if they were aware of a larger pattern, they can't account for the pattern. It seems to be random, although there must be some hidden connection. 

But their inability to identify the collusion in no way obviates the evidence of cheating in the individual cases. By the same token, even if the distribution of special providences appears to be random, that doesn't affect or cancel out the evidence in specific cases.

12 comments:

  1. I'm not sure how to phrase the question, but at what point does a casino suspect cheating? Is there a threshold or tipping point beyond which a winning streak cannot be accounted for by luck? If so, what is it? Can we as Christians use this same idea to point out that evolution requires just to many "coincidences" and really there must be an intelligent designer behind it? Is there a succinct way to make this point in a sound byte, or does it require a long explanation?

    ReplyDelete
    Replies
    1. http://www.uncommondescent.com/intelligent-design/coordinated-complexity-the-key-to-refuting-postdiction-and-single-target-objections/

      Delete
  2. The thing is though, your explanation is equally compatible with astrology, casting chicken bones, or prayers to Ganesh. A larger sample might reveal that any one of those things might be actually causing a currently non-discernible pattern. If you cannot identify the pattern, you cannot credit its cause. Therefore, you are left with indistinguishable noise on the one hand or wildly unjustified claims on the other.

    >“By the same token, even if the distribution of special providences appears to be random, that doesn't affect or cancel out the evidence in specific cases.”<

    If you cannot distinguish that there is a pattern to begin with, you have no evidence of specific cases. The existence of the pattern is what gives you the “special cases.”

    Your argument is akin to walking along a beach and saying that it is really part of a continent wide sand painting of the Mona Lisa and you just so happen to be walking on the natural-colored part and therefore this might be evidence of da Vinci writ large. However, without the satellite data, you have no more reason to believe that claim than the claim that it is a Salvador Dali or Jackson Pollock. Indeed, you haven’t established why anyone should take your claim seriously at all that it is anything more than a beach.

    Your group of gamblers analogy is inapplicable to what you hope to demonstrate because you know from the start that there is a group of gamblers and know therefore what they are doing. (And it is also naturalistic in origin.)

    What you hope to demonstrate is that without knowledge of a pattern or even its possibility of discernment, that you can then use that to validate specific “special” instances of that pattern. Without the pattern, you do not know what you have is a “special instance.”

    ReplyDelete
    Replies
    1. "The thing is though, your explanation is equally compatible with astrology, casting chicken bones, or prayers to Ganesh."

      Only if those were too accurate, naturally improbable, and opportune to be sheer coincidence. It depends on the specific example.

      "A larger sample might reveal that any one of those things might be actually causing a currently non-discernible pattern. If you cannot identify the pattern, you cannot credit its cause. Therefore, you are left with indistinguishable noise on the one hand or wildly unjustified claims on the other."

      i) You suffer from the peculiar, self-defeating notion that we can't understand the specific unless we understand the general. That's backwards. Our understanding of the general is an abstraction from specific case.

      ii) Likewise, in my gambling analogy, the casino can know there's a suspicious cause even if it can't say how it's done.

      "If you cannot distinguish that there is a pattern to begin with, you have no evidence of specific cases. The existence of the pattern is what gives you the 'special cases.'"

      You're repeating the same mistaken. It's unnecessary to begin with a pattern to identify specific cases. The latter can have their own individual evidence. That's why I use the gambling analogy. It's unnecessary to first establish a general pattern of cheating at several casinos to establish cheating at any particular casino. And in my illustration, the casino can't even prove cheating, beyond the statistical aberrations.

      "Your argument is akin to walking along a beach and saying that it is really part of a continent wide sand painting of the Mona Lisa and you just so happen to be walking on the natural-colored part and therefore this might be evidence of da Vinci writ large."

      Your counterargument is akin to saying we can never establish cheating.

      "Indeed, you haven’t established why anyone should take your claim seriously at all that it is anything more than a beach."

      i) You don't speak for anybody besides yourself.

      ii) My claim establishes a principle. You're confusing that with whether specific examples are probative.

      "Your group of gamblers analogy is inapplicable to what you hope to demonstrate because you know from the start that there is a group of gamblers and know therefore what they are doing."

      The casino doesn't have to know from the start that there's a group of gamblers to figure out that someone is cheating.

      "(And it is also naturalistic in origin.)"

      Irrelevant to the principle.

      "What you hope to demonstrate is that without knowledge of a pattern or even its possibility of discernment, that you can then use that to validate specific “special” instances of that pattern. Without the pattern, you do not know what you have is a 'special instance.'"

      I suspect one of your confusions is your equivocation over the notion of a "pattern". In my illustration, there can be a discernible pattern of cheating at any particular casino without anyone having to know that that's part of a larger pattern of cheating at multiple casinos. They don't need a "pattern" in the sense of background information about a cheating ring to detect a pattern in the sense of a player with a winning streak that's too improbable to be dumb luck.

      Likewise, one can detect cheating without being able to detect the method.

      By the same token, there can be verifiable cases of special providence without knowing why that's not more frequent or equitable.

      Delete
  3. >”You suffer from the peculiar, self-defeating notion that we can't understand the specific unless we understand the general.”<

    No, your argument requires that the general must be known in order to identify the specific. Without knowing the overall larger pattern, your claims of special providence are exactly on par with any other number of superstitions.

    >”Your counterargument is akin to saying we can never establish cheating.’<

    No, I’m saying that without the larger pattern, you cannot establish the concerted action of a group of cheaters. We are not arguing about whether cheating is detectable; we are arguing about whether the actions of a supernatural agent are detectable as part of a pattern—analogous to the concerted group action in your example.

    The flaw in your argument rests right there. You are making the unjustified grasp from the general to identify your specific and you don’t have the general. Identifying your claims of specific providence as indeed providence only works if you can identify a general pattern of providence and situate them therein. Your claims of special providence in the specific necessarily hinge on being able to identify them as part of a general divine pattern. You concede that discerning the general pattern is not possible; therefore your argument fails.

    Without the discernible larger pattern, your claims of specific providence are indistinguishable from expected naturalistic distribution of events on the one hand or open superstition on the other if we accede supernatural causation.

    In short, without the general pattern you have no justification or reference with which to label events as specific providence. Or, to use your analogy, you are unjustified in saying that cheating in one casino is part of a conspiracy if you concede from the outset you have no way of knowing that a conspiracy exists.

    ReplyDelete
    Replies
    1. "No, your argument requires that the general must be known in order to identify the specific. Without knowing the overall larger pattern, your claims of special providence are exactly on par with any other number of superstitions."

      You're repeating your original contention without advancing the argument. I explained to you why that's false.

      "No, I’m saying that without the larger pattern, you cannot establish the concerted action of a group of cheaters. We are not arguing about whether cheating is detectable; we are arguing about whether the actions of a supernatural agent are detectable as part of a pattern—analogous to the concerted group action in your example."

      You're still confused. My argument doesn't require the casino to establish the concerted action of a group of cheaters in order to identify particular instances of cheating. Indeed, my point was just the opposite: any given casino could be in the dark regarding a cheating-ring. They don't need to know that to know that a player at their casino is cheating.

      Moreover, they might infer that this must be the result of collaboration without ever knowing the identity of the collaborators.

      Likewise, my argument wasn't predicated on whether the actions of a supernatural agent are detectable as part of a pattern. To the contrary, I argued that they can be detectable in ignorance of a larger pattern. That's the actual function of my cheating-ring analogy.

      You seem to think that because the reader is privy to insider information vis-a-vis my analogy, that probative evidence for examples of special providence depends on having that context. Wrong. That's just an illustration to give an overview of the issues, and distinguish one perspective from another.

      "The flaw in your argument rests right there. You are making the unjustified grasp from the general to identify your specific and you don’t have the general."

      That's your persistent confusion. I'm not arguing from the general to the specific. The general wasn't by starting-point.

      "Identifying your claims of specific providence as indeed providence only works if you can identify a general pattern of providence and situate them therein."

      They don't have to be classified as instances of special providence to be recognizes as instances of supernatural agency. "Special providence" is just a handy label for varied phenomena of a certain kind.

      "Your claims of special providence in the specific necessarily hinge on being able to identify them as part of a general divine pattern."

      You keep asserting that when the structure of my argument was precisely the opposite.

      "Without the discernible larger pattern, your claims of specific providence are indistinguishable from expected naturalistic distribution of events on the one hand or open superstition on the other if we accede supernatural causation."

      By that logic, you could never distinguish a coincidence from cheating.

      "or reference with which to label events as specific providence."

      I don't commence with that reference class. Rather, that's the result of induction.

      "Or, to use your analogy, you are unjustified in saying that cheating in one casino is part of a conspiracy if you concede from the outset you have no way of knowing that a conspiracy exists."

      Irrelevant. The casino can be ignorant of a conspiracy. That does nothing to obviate evidence of cheating in any particular instance.

      Likewise, there can be evidence of supernatural agency absent knowledge of a general pattern.

      You seem to be hung up on the label: "special providence". But that category does no work in the argument. It's just a convenient designation.

      Delete
  4. >"Moreover, they might infer that this must be the result of collaboration without ever knowing the identity of the collaborators."<

    This is precisely one of my points. To use your analogy, the collaborators may include Batman, Superman, or the Ku Klux Klan--the casino has no way of knowing.

    And relating this back to your larger argument, you are still stuck on one of the horns I identified originally and have continued to assert.

    Arguendo, conceding that you do indeed have an incident that justifies even a provisional classification of special providence or supernatural agency, without the general pattern with which to identify the source of the answered prayer of miracle, you are no more justified in attributing it to the Christian God than to any open class of superstitions or religions. The fact is, this type of “evidence” you claim is also claimed by religious practitioners the world over for their particular brand of religion and by any number of believers in pseudoscience and superstition.

    Evidence that works equally well for Ganesh, astrology, and chicken bones cannot be honestly used for evidence of the Christian God and is, indeed, no evidence at all.

    ReplyDelete
    Replies
    1. "We are not arguing about whether cheating is detectable; we are arguing about whether the actions of a supernatural agent are detectable…"

      That may not be what you're attempting to argue, but that's my argument, where detecting supernatural agency is analogous to detecting casino cheaters.

      "…without the general pattern with which to identify the source of the answered prayer of miracle, you are no more justified in attributing it to the Christian God than to any open class of superstitions or religions."

      I don't need a general pattern to ID the source. To paraphrase Mackie, a miracle is an event that would not have happened if the natural world was left to itself, as opposed to outside agency (i.e. supernatural intervention).

      Since he's an atheist, I'm not tilting the scales in favor of theism or Christian theism in particular by using his definition.

      i) On that definition, I don't need a general pattern to ID a supernatural source. Rather, in case the incident would not have happened if nature was left to run its course, then by process of elimination, the cause must be supernatural.

      ii) Even if (ex hypothesi) it doesn't select for Christian theism in particular, it rules out naturalism and reduces the remaining explanatory options.

      We don't need a silver bullet. There are multiple lines of evidence for Christian theism.

      iii) It also depends on the specific context, viz. is this an answer to Christian prayer?

      "Evidence that works equally well for Ganesh, astrology, and chicken bones…"

      Before you can even get to that point, you'd need to establish an event under those auspices that fits the definition and meets the criteria (a la Mackie).

      Postulating hypothetical undercutters doesn't prove or disprove anything. There's concrete evidence for miracles in a Christian context. Well-documented cases.

      Atheists have a bad habit of retreating into safe abstractions rather than grappling with specific evidence in specific cases.

      Delete
  5. > “but that's my argument, where detecting supernatural agency is analogous to detecting casino cheaters.”<

    Overall your analogy is pretty good for your argument but it is at this point on which it suffers.

    Cheaters in a casino are a natural phenomenon. Therefore, we are getting tangled up on some misunderstandings of statistics. Events that are not probable (winning cheaters) are not justifiably analogized to (impossible by natural means) miracles.

    Events that by happenstance match the Yes/No/Maybe/Later answered prayer template can also match the +/-/null (human perception) natural flow of events (hence being of the Post hoc ergo propter hoc variety of superstition). Your argument rests on this perception/interpretation.

    Replace a Muslim today in your shoes and whatever incident you point to, he would point to as evidence of Allah’s will in line with Islamic theology. Replace a Greek of millennia ago in your shoes and get their gods as answers. Etc. You have no criteria with which to identify the object of your argument: identifying fortuitous events as the agency of the Christian God. In the interpretation of any other believer these evidences are for any of their beliefs.

    >”There's concrete evidence for miracles in a Christian context. Well-documented cases.”<

    So your argument (improbable events = evidence of Christian theism) ultimately rests on believing that Christian theism is true. Your argument is tautological in nature the way you have expressed it.

    An honest rendition of your argument would merely be: “Assuming Christian theism, I can interpret improbable events as the will of God, providence, answered prayers, etc.”

    If you would have expressed it that way from the beginning, there would be no problem and I wouldn’t even have been moved to point out your current error as there would have been none committed.

    >”Atheists have a bad habit of retreating into safe abstractions rather than grappling with specific evidence in specific cases.”<

    This is a puzzling statement given the context; your entire argument is a hypothetical and contains no specific cases.

    ReplyDelete
    Replies
    1. "Cheaters in a casino are a natural phenomenon."

      i) A red herring. The question at issue is the general principle: can we identify specific examples of personal agency even if we can't identify an overall pattern?

      ii) In addition, an analogy will typically use an example that's less contested than the thing that it's employed to illustrate.

      "Therefore, we are getting tangled up on some misunderstandings of statistics. Events that are not probable (winning cheaters) are not justifiably analogized to (impossible by natural means) miracles."

      Another example of your inability to reason clearly. What are the odds of winning a poker hand? Well, poker is, in part, a game of chance (as well as skill). The deck is supposed to be randomly shuffled. That's analogous to letting nature run its course.

      Cheating changes the odds because it overrides the random element. That can involve a stacked deck. Or, in my illustration, the cheaters know what cards each player has. That's analogous to an external agent who manipulated the outcome. You get a different result than if nature was left to its own devices.

      Put another way, it's the difference between a closed system and an open system. A game of chance is like a closed system. Cheating is like an open system.

      Outside intervention changes the way nature behaves. That's true with respect to cheating and miracles alike.

      "Events that by happenstance match the Yes/No/Maybe/Later answered prayer template can also match the +/-/null (human perception) natural flow of events (hence being of the Post hoc ergo propter hoc variety of superstition). Your argument rests on this perception/interpretation."

      Is the casino guilty of the post hoc ergo propter hoc fallacy when it suspects cheating? How do you distinguish cheating from coincidence?

      "Replace a Muslim today in your shoes and whatever incident you point to, he would point to as evidence of Allah’s will in line with Islamic theology."

      i) No, a Muslim wouldn't point to a prayer to the Trinitarian God or the Son of God (Jesus) as evidence of Allah's will in line with Islamic theology.

      ii) Furthermore, creating imaginary parallels between Muslim answers to prayer and Christian answers to prayer is just that…imaginary. That isn't evidence because that isn't real. There's nothing to evaluate.

      To even get off the ground, it's necessary for you to compare and contrast specific cases and specific evidence. There are entire monographs on well-documented Christian miracles. For instance, take Craig Keener's two-volume study on Miracles: The Credibility of the New Testament Accounts. Or Rex Gardner, Healing Miracles: A Doctor Investigates. You need to come up with something on the same plane from a Muslim standpoint to even initiate a comparison. You can't just wing it with abstractions.

      Delete
    2. iii) Your attempted comparison backfires. Consider literature like Nabeel Qureshi, Seeking Allah, Finding Jesus: A Devout Muslim Encounters Christianity; David Garrison, A Wind In The House Of Islam: How God Is Drawing Muslims Around The World To Faith In Jesus Christ;Tom Doyle, Dreams and Visions: Is Jesus Awakening the Muslim World?

      These involve Muslims who attest supernatural experiences that drew them to Christianity.

      iv) Christian theism doesn't preclude God answering the prayer of a Muslim. For instance, if God intends Nabeel Qureshi to become a Christian apologist, God might answer the prayer of Nabeel's Muslim grandfather if Nabeel's future existence depended on it.

      v) Muhammad sabotaged his prophetic claims when he made the Bible the standard of comparison. Measured by his own yardstick, he comes up short. Islam and Christianity are asymmetrical: the veracity of Islam is contingent on its agreement with the Bible; the veracity of Christianity is not contingent on its agreement with the Koran.

      "Replace a Greek of millennia ago in your shoes and get their gods as answers. Etc."

      That's not a counterexample to my position because there's nothing concrete to assess. You're not presenting actual examples.

      "You have no criteria with which to identify the object of your argument: identifying fortuitous events…"

      I'm not appealing to fortuitous events–just the opposite.

      Delete
    3. "…as the agency of the Christian God."

      I already addressed that objection in a previous comment. You're repeating yourself without engaging the response.

      "In the interpretation of any other believer these evidences are for any of their beliefs."

      You're speaking in fact-free generalities.

      "So your argument (improbable events = evidence of Christian theism) ultimately rests on believing that Christian theism is true."

      i) One of your persistent problems is that you're attacking an argument I never made. The topic of my post wasn't evidence for Christian theism, per se, but evidence for God's existence–in contrast to atheism.

      ii) My argument wasn't based on "improbable" events. That oversimplifies what I said. Rather, I said events "too specific, unlikely, and opportune to be sheer luck."

      iii) Your inference is fallacious. My argument doesn't rest on believing Christian theism is true. Rather, I alluded to evidence for Christian miracles. Hence, your next two sentences piggyback on your invalid inference.

      "This is a puzzling statement given the context; your entire argument is a hypothetical and contains no specific cases."

      That's because you're a Johnny-come-lately. I've been posting material on miracles for a dozen years.

      Delete