ASA Connect

 View Only
Expand all | Collapse all

The Nature of Reality

  • 1.  The Nature of Reality

    Posted 01-11-2018 21:13

    I would like to share, with all who read this, my view of "the nature of reality", at which point I would appreciate your opinions.  I will begin by pointing out what I believe to be a logical flaw in the movie "Good Will Hunting".

    Specifically, in the scene where psychologist Dr. Sean Maguire (Robin Williams) tells Will (Matt Damon) about the first time he met his wife, there seems to be an implied assumption that if Sean had gone to "the game" (Game 6 of the World Series in 1975), instead of staying at the bar where he had just met his future wife, then the very famous home run hit by Carlton Fisk would still have occurred.  I contend that if Sean had gone to the game, the game would have played out completely differently, and the famous home run which actually occurred would not have occurred – that's not to say that some other famous home run could not have occurred.  It seems to be clear that neither characters Sean nor Will understand this – and I contend these two supposedly brilliant people would have known better!  It is certainly clear that neither Matt Damon nor Ben Affleck (the writers) understand this.

    Along the same lines, I think it's a good thing that I chose not to go to the Cal-Stanford football game in 1982 (a friend had an extra ticket), because if I had gone, then "The Play" would have never occurred!

    I have run my ideas regarding this situation (and more generally my understanding of the nature of reality) by some of my friends, colleagues, and students, and for the most part they think I'm a dope!  One exception was a 14 year-old (at the time) young man (son of an astronomy professor at my university) who correctly summed things up with the statement "everything that happens is a fluke".

     

    So what do you think?

     

    Dr. Laurence D. Robinson

    Associate Professor of Statistics

    Department of Mathematics and Statistics

    Ohio Northern University

    Ada, OH 45810

     

    (419) 772-2358

    L-Robinson.1@onu.edu

     

     

    From "Good Will Hunting"

    https://www.youtube.com/watch?v=jg_9FQk6UnA

     

    "The Play"

    https://www.youtube.com/watch?v=mfebpLfAt8g



  • 2.  RE: The Nature of Reality

    Posted 01-12-2018 06:44
    I want to mention that, secretly, I went to that game, and that is why there was a home run. Actually, it was not me. I can not claim credit. Before the game I bought a cheese sandwich, and, while they were putting the sandwich together, they were putting only two pickle slices on top of the cheese. I asked them to put an extra pickle slice, which they willingly did. I do believe that it was the extra pickle slice that caused the home run. Alternatively, it may have been the red shirt that I was wearing. I have noticed that good things happen when I wear that red shirt. That is far as I can take it. If we wade into the real meaning of life and the nature of Reality (capital "R"), we might never finish this discussion. 

    Postscript: you are definitely not a dope, and the 14-year gave an excellent 14-year-old answer. Let us see what the 14-year-old thinks in another decade or so. And, I would recommend asking a 7-year-old. They have a much more fun interpretation of Reality than we do. 

    Thanks for the opportunity to have some fun.

    Nayak




    ------------------------------
    Nayak Polissar
    Principal Statistician
    The Mountain-Whisper-Light Statistics
    ------------------------------



  • 3.  RE: The Nature of Reality

    Posted 01-12-2018 09:19
    What you have mentioned is actually called "counterfactual theory of causality" and has a lot of support  in philosophical and  statistical literature (Causality: Models, Reasoning and Inference), as well as the strong opposition (what is, specifically, my view Troublesome Dependency Modeling: Causality, Inference, Statistical Learning by Igor Mandel :: SSRN). So, I'm on your side - if something had not happened, no future event depending on that will happen with certainty, the picture of the world will be infinitely changed if this is what you mean.

    ------------------------------
    Igor Mandel
    Telmar, Inc.
    ------------------------------



  • 4.  RE: The Nature of Reality

    Posted 01-12-2018 09:22

    I mostly agree with what you are saying, but there is an empirical question involved.

     

    Had the order of the starting players been changed, I have little doubt that the accumulated small changes from that would mean that most of the later details of the game would be different. The butterfly wing phenomenon.

     

    But a single individual in the stands? Not sure. The butterfly's wing may change the patterns enough as to affect storm sequences in 1,000 years, but it doesn't do so immediately.  So whether one person would be enough in that short time frame to make a difference is unclear.

     

    A related note is that if a couple who would have had a child tonight is interrupted and instead conceives a child tomorrow night, the effect on future humanity is huge and cumulative.  Almost certainly a different sperm will get through, hence a different child. That child's friends and eventual relationships and offspring will all be different from what would otherwise have happened. Plus whoever marries that child will not have married the person they would have married, creating a ripple effect. Friends will be introduced differently, creating still more different relationships. Plus many of the children of couples who still form in the new situation will be conceived at different times than they otherwise would have, due to the addition of new friends and relationships, hence they will be different as well. In a few generations large numbers of living people are different individuals than would have been in the original case! It's not better or worse, just different.

     

    Ed






  • 5.  RE: The Nature of Reality

    Posted 01-15-2018 20:27
    Not only that Ed.  It is well known that having children is hereditary. That is, if your parents did not have children, it is certain that you won't either.

    ------------------------------
    Jerome Yurow
    Retired
    ------------------------------



  • 6.  RE: The Nature of Reality

    Posted 01-12-2018 09:32

    greetings Dr. Robinson,

     

    here's my problem with your position:  the use of the word would vs. could – when you say " I contend that if Sean had gone to the game, the game would have played out completely differently" where is your evidence that that particular event could not happen, you are suggesting that  Sean WOULD have some causal effect on the batter, but why? if you change your assertion to "I contend that if Sean had gone to the game, the game could have played out completely differently" then I have no problem with what you suggest --

     

    thanks,

     

    cheers,

    joe

     






  • 7.  RE: The Nature of Reality

    Posted 01-12-2018 12:45
    Is everyone except Gracely and Visconi being sarcastic or do the others really believe that the presence or absence of a single fan unknown to the players can have any effect on the game?

    ------------------------------
    Emil M Friedman, PhD
    emilfriedman@gmail.com
    http://www.statisticalconsulting.org
    ------------------------------



  • 8.  RE: The Nature of Reality

    Posted 01-14-2018 18:02
    The problem with you proposition is that you do not consider probability and agency. If any member of the general public goes or does not go to a game, it is unlikely to affect the game. It is highly unlikely that the character going to the game would have resulted in the homerun not happening. Going or not going to a game would result in a change in reality, but why would it result in a homerun not happening? It might affect the significant others of the character and others that he comes into contact with. It would take a series of unlikely events for it to affect the game. For example, X goes to the game by car. This extra car on the street affects a player's wife who does not make it to the game. This affects how one player plays which affects the other players. Hence, no home run. This is an unlikely scenario. For any one person's (excluding the powerful) trivial behavior to have such major consequences is unlikely to happen in the short run. Yes, in the long run it could have major consequences. What makes it even more unlikely is that you are referring to one specific event out of huge number of events. Why would that specific event change and not another event?

    I hope I am being clear. 

    Michael

    ------------------------------
    Michael Maranda, PhD
    ------------------------------



  • 9.  RE: The Nature of Reality

    Posted 01-15-2018 14:30
    Michael, you and I agree.  I was just wondering whether someone saying, "The home run occurred (or didn't occur) because I went (or did not go) to the game." was being facetious rather than being honestly foolish.

    ------------------------------
    Emil M Friedman, PhD
    emilfriedman@gmail.com
    http://www.statisticalconsulting.org
    ------------------------------



  • 10.  RE: The Nature of Reality

    Posted 01-15-2018 16:53
    For the uninitiated, Dr. Robinson et alia are referred to the short story by Ray Bradbury, A Sound of Thunder, in which a hunting party travels back in time to a prehistoric past that results in a disastrous change in the future when the party returns.  Of course, it was fiction, but is consonant with the present day butterfly catastrophe theory, which is also something of a fiction.

    More than 30 years ago, Paul Holland published an article in JASA entitled "Statistics and Causal Inference" [JASA 1986, Vol. 81: 945 ff] in which he quite clearly stated the problem associated with counterfactual arguments.

    Solipsism should be left to the millenials.  Dr. Robinson would do well to listen to his associates.

    Gene S. Fisch,
    Paul Chook Dept. of IS & Statistics
    CUNY/Baruch College

    ------------------------------
    Gene S. Fisch
    Paul Chook Dept. of IS & Statistics
    CUNY/Baruch College
    ------------------------------



  • 11.  RE: The Nature of Reality

    Posted 01-16-2018 22:16
    How do you know?

    Jonathan Siegel
    Associate Director Clinical Statistics

    Sent from my iPhone




  • 12.  RE: The Nature of Reality

    Posted 01-17-2018 12:37
    I agree with Jonathan on this. Yes, there is no proof that the home run is 100% guaranteed to happen. But that does NOT imply that there is proof that it will not happen. How many times has industry produced a new chemical with no proof of harm - however, when it is in use with either much greater or longer exposure than the prerelease testing encompassed, harm is revealed?

    Sent from my iPhone




  • 13.  RE: The Nature of Reality

    Posted 01-18-2018 04:48
    What Ellen Barnes said. There is an old saying among logicians: "Absence of evidence is not evidence of absence".





  • 14.  RE: The Nature of Reality

    Posted 01-19-2018 11:35

    Philosophers are notably adept at inventing imaginative and amusing counterfactual scenarios for questions like the one that started this thread. I went through a number myself and then decided to winnow them by challenging myself to find the least improbable physically/psychologically possible counterfactual that could account for the attendance of just one more fan at the ballgame causing the home run not to be hit. And since this site is not devoted simply to creative fiction writing I challenged myself to describe how the proposed counterfactual would have any relevance to statistics. So here goes.

    Professional athletes have trained themselves to block out crowd noise. Yet the capacity to drown out crowd noise is subject to threshold limits. There is a decibel level above which even the athlete with the greatest powers of concentration can no longer perform optimally. Thus it is possible (though highly unlikely) the presence of one more screaming fan could raise the decibel level just enough to exceed the batters ability to fully concentrate and he fails to hit the home run.

    Is there any statistical significance to this counterfactual scenario? If sports statisticians were to seek to model the probability of the batter hitting a home run they would likely develop a regression model specifying a variety of independent variables (the batter's historical rate of home runs per at bat, the handedness of the pitcher, the number of men on base, would all be likely candidates with coefficients unlikely to equal zero). Would we choose a predictive model that included crowd size and noise as a regressor? Somewhat facetiously I would suggest that crowd size would likely be embedded in the "e" term of any model, the "e" term sometimes is of course referred to as a "noise" factor.

    Interestingly sports handicappers in football have long factored in crowd noise in the predicted outcome of games under the heading of home field advantage. The crowd becomes the twelfth player. It is a common sight to see the home team's bench exhorting the crowd to yell louder on third down plays to make it harder for the visiting team's offense to hear their quarterback counting down to the snap.

    The kind of threshold condition in my counterfactual occurs all the time in nature and when a stable system reaches the critical state just one more of something (picture straws and camels) creates dramatic changes. In non-critical states small changes in causal variables lead to only small effects. When systems become unstable a small change in a causal variable leads to a large effect.  But the modeling of thresholds phenomena utilizes powers laws, regression modeling on the other hand is an averaging or smoothing process. The latter will yield decent prediction results when things are stable but things do not always remain stable. Creating predictive models that factor in the possibility of very rare but destabilizing events is definitely a topic of statistical interest.  



    ------------------------------
    Michael Sack Elmaleh
    Principal
    Michael Sack Elmaleh CPA, CVA
    ------------------------------



  • 15.  RE: The Nature of Reality

    Posted 01-19-2018 12:46

    Hi, Michael:


    Your argument is interesting, but I think you are focusing too much on the meaningful predictability of the result, which misses the flavor of this idea.


    Imagine that one more person is in the stands, and the batter who is third before the big home run happens to notice something because of it. That batter takes a second longer to get to the plate. Tiny change with no expected direction of effect. But it could still cause a divergence of reality. A second later means that both batter and pitcher are slightly different in position and motion. The pitches that come may be slightly different from what would have come in the original scenario. The result may be slightly different, or may cross a threshold (ball hit/not hit). 


    Once even a minor threshold like that is crossed, even if the batter ultimately ends up at the same base, now we are 10, 20, 30 seconds and a different game sequence different. Everyone is thinking a bit differently.


    This can affect what happens to the next batter, and so on.


    Does the initial one-second delay mean that we should expect more or fewer home runs? No!  But if we knew there was a big home run in the original scenario, the one second delay makes it less likely that specific one would occur. It occurred because of the exact specific conditions, down to split second timing and attitudes, present that day. Any changes in those would likely have prevented it.


    Again, I can't say how likely this pattern is -- it's too complicated to assess with any measurements we now have. But it's interesting the ponder!


    Ed


    Ed J. Gracely, PhD
    Associate Professor
    Family, Community, & Preventive Medicine

    College of Medicine

    Associate Professor

    Epidemiology and Biostatistics

    Dornsife School of Public Health

    Drexel University
    2900 W. Queen Lane,
    Philadelphia PA, 19129

    Tel: 215.991.8466 
    | Fax: 215.843.6028
    Cell: 609.707.6965

    eg26@drexel.edu (egracely@drexelmed.edu forwards)
    drexelmed.edu  |  drexel.edu/publichealth






  • 16.  RE: The Nature of Reality

    Posted 01-19-2018 19:08
    Dear Ed, 

    The whole point is that a little change is not likely to result in any big difference in the short run. Much less one specific event changing.

    Any change would be a change in reality. (Of course this can not happen because once an action happens it is fixed in the reality of our universe.) But, a trivial change will not likely result in significant change in the short run, especially when it is not directly related to the event in question. I believe that the confusion stems from how chaos theory is misrepresented with the butterfly example and how the multiverse theory is misrepresented.  Also, postmodernism probably has something to do with this confusion.

    The point of the short story by Bradberry and the butterfly example (which is a misrepresentation of chaos theory) is that a small trivial change in the present will result in big changes in the long run. What the changes there will be are not specified.

    Your argument is flawed. It is based on a whole series of improbable events that result in one specific event. You wrote "notice somewhat". What would he notice? Is the batter capable of noticing one extra person in the stands? Although I do not watch sports I know that the other team tries to distract the batter. A professional ball player would have enough concentration not to be distracted by one extra fan assuming he could tell the difference. Most likely he could not tell the difference in the number of fans. I have not seen a ball game in years, but I know that the pitcher and the batter do a number of things before the ball is thrown. And, there is no evidence that these small changes you mentioned would result in anything if they happen. More likely they would not happen. Even if they happened the pitcher would assume the proper position and the batter would do what he does at the plate. They would do what they are trained to do, what they have practiced. Yes, one or both can have an off day, but not because there is an extra anonymous fan.

    Bassically you repeat my argument that a series of highly improbable events has to happen to change this one event and come to the opposite conclusion that this is reasonable scenario. 

    To answer another person's posting, home field advantage does not refer to the noise of the crowd. It refers to the support for the home team and the hostility to the visitors. And, our perception of loudness is dependant on a number of factors. The perception of loudness though (assuming all things being equal, which they never are) is based on a logarithmic scale. It is almost beyond imagining that one fan would change a player's perception of loudness.

    No I do not think I miss the flavor of original posting. The post mention the negative reaction that he received from friends and colleagues. I think that the premise was based on highly flawed speculation. It gave too much power to human agency and distorts how to handle counterfactuals. When dealing with counterfactuals we do not assume that highly improbable events will happen. That would make no sense. We speculate about what is likely to happen. The reason we do randomized control experiments is to deal with counterfactuals. In a Saturday Night live sketch there was speculation about how would it affect World War II if Eleanor Roosevelt could fly. That sketch was somewhat funny. This discussion is sad. Sillness like this on a scholarly forum can have negative consequences. What are we saying about counterfactuals? That scholars/statisticians/researchers can dream up improbable scenarios?

    I decided not to have Greek lemon potatoes at lunch. Will that decision affect the super bowl or nuclear war with North Korea? It did affect my blood sugar. 

    I have a confession--I enjoy Doctor Who and I love science fiction. But, I know the difference between science fact and fiction.

    I am disappointed we wasted so much time on this when there are so many important things that we can discuss.

    Best wishes.

    Michael
    Michael J. Maranda, PhD





  • 17.  RE: The Nature of Reality

    Posted 01-20-2018 14:00

    Thank you, Michael.


    gsf





  • 18.  RE: The Nature of Reality

    Posted 01-20-2018 15:10

    I was gonna let this die, but seeing someone agreeing with Michael, I won't.


    My argument does NOT depend on extremely unlikely counterfactuals. It assumes instead that once even one slight change happens, that the effect on reality is cumulative in basically random ways that grow exponentially in impact.


    I like my example of the effect of a timing change on a pregnancy because you can more easily see the cumulation. Someone might argue that a 10-minute delay in when a couple has sex is of no consequence and will have 0 effect.  Any slight effects will soon average out. 


    But that couldn't be further from the truth. Yes, the average number of males and females, tall and short, smart and slow kids born over the next few generations will be the same (within randomness), But the specific individuals may all be different (at least within a circle of affect). 10 minutes later --> different sperm impregnates --> different child --> other children of the couple different --> different friends, next generation in their circle, etc... It grows rapidly to change a large number of people, in random ways.


    I said in my original post that whether someone in the stands would be enough change to cause a sequence that would change the bit home run is unclear. I might agree it was unlikely.  But what about something a bit more specific and visible, like a dog running across the field?  It has no real consequence for anybody's playing ability, BUT would the pitcher who threw the homerun pitch and hitter who hit it be in exactly the same positions that allowed that rare event half an hour later as if there had been no dog?  I would say not. The dog does not in the long run affect how many home runs are hit. But it may change when and where they are hit, in random fashion.


    Ed




    Ed J. Gracely, PhD
    Associate Professor
    Family, Community, & Preventive Medicine

    College of Medicine

    Associate Professor

    Epidemiology and Biostatistics

    Dornsife School of Public Health

    Drexel University
    2900 W. Queen Lane,
    Philadelphia PA, 19129

    Tel: 215.991.8466 
    | Fax: 215.843.6028
    Cell: 609.707.6965

    eg26@drexel.edu (egracely@drexelmed.edu forwards)
    drexelmed.edu  |  drexel.edu/publichealth






  • 19.  RE: The Nature of Reality

    Posted 01-20-2018 20:32
    Ed,

    I hope you understand the difference between oranges and apples.

    Your example from your first post consisted of improbable events. First you deny it and then you admit it. You then offer another scararo that has nothing to do with an anonymous fan. Yes, if a dog ran on the field it might affect the game, but then again it might not. But, why would that happen? It is not relevant to original scenario. Yes, if an event occured that delayed the playing of the game--then events after this incident would occur at a different time. But, this is true by definition.

    Have you ever been to a sporting event? All kinds of things happen. At the only hockey game I went to they took a fan away because he got a puck in the mouth. Do you understand how athletes train? Do you know that they need concentration? That they learn to concentrate on their performance despite distractions including pain? They are down by three points and their best player is injured or they are up by five points. All kinds of potentially stress things happen. No two games are exactly alike. Every game is a combination skill and luck.

    Your example about pregnancy is of a different order than the baseball game problem. First you are talking about a direct action of an actor involved in the relevant outcome. Second, you are talking about long term effects. Although there would be a short term effect--a different child. This of course assumes that a delay of ten minutes would affect which sperm fertilizes the egg. I do not know enough about the biology of this to say that this is a likely change. It could be that it would not make a difference. Did you research this to determine if it would have an effect? Does the literature support your proposition? Do sperm change their position over a short time frame? Yes, my not eating Greek lemon potatoes had an effect on my blood sugar levels in the short term, but it will have no effect on the super bowl. It is even unclear if this one event will have a long term effect on my health.

    The belief that one's attendance or lack of attendance at a large scale sporting event (assuming you are anonymous fan) will have any effect on the game or specifically its outcome is superstition. It clearly meets the definition of superstition. I always try to remember to take a book with me to medical appointments because when I don't I have to wait a long time. So, my taking a book it affects my waiting time. This is my personal superstition. But, it is my superstition. And, for one to believe that their attendance or absence at a large professional sporting event where they are anonymous fan has an effect on the game is egotistical at best. 

    Ed why don't you propose a more serious question?

    Michael










  • 20.  RE: The Nature of Reality

    Posted 01-20-2018 21:43

    Michael, you talk about apples and oranges. So let me propose two.


    Apple: The dog walking across the field makes a predictable change in the outcome. Because of it, team A is now more likely to win.  Or, maybe, because of it, the teams will both play noticeably less well, and the game will deteriorate.


    Orange: The dog walking across the field will produce tiny, cumulative, but random changes in the game.  The precise split second decisions that led to the ball being in exactly the right place to meet the bat in exactly the right place for a home run rather than a base hit in the first inning will change. It is just as likely that the conditions will change to produce a home run as to prevent one -- it's random. However, as these random changes happen, they cumulate. Even one change such as base hit versus home run may affect how many are on the bases, how many batters come up etc.


    The apple belief is superstition. I fully agree.


    The orange belief is not, because I'm not predicting anything systematic.


    What the orange belief does predict is that if the dog walks across the field, the game will play out differently.  So in the present reality, if some odd combination of events produced a special outcome, like a bases-loaded home run, it is very unlikely that exact event will happen again in the alternative counterfactual reality of the dog walking.


    However, sight unseen it is just as likely for such an event to occur with or without the dog walking.


    I'm happy to continue the discussion, but I am finding your tone a bit condescending. If it remains so, I shall delete rather than respond to future messages from you!


    Ed


    Ed J. Gracely, PhD
    Associate Professor
    Family, Community, & Preventive Medicine

    College of Medicine

    Associate Professor

    Epidemiology and Biostatistics

    Dornsife School of Public Health

    Drexel University
    2900 W. Queen Lane,
    Philadelphia PA, 19129

    Tel: 215.991.8466 
    | Fax: 215.843.6028
    Cell: 609.707.6965

    eg26@drexel.edu (egracely@drexelmed.edu forwards)
    drexelmed.edu  |  drexel.edu/publichealth






  • 21.  RE: The Nature of Reality

    Posted 01-21-2018 01:35
    Ed,

    Of course you can make up any scenario you wish. These are irrelevant to the original problem which was whether the Robin Williams's character attending or not attending the game would have affected the home run. You wrote:

     The dog walking across the field will produce tiny, cumulative, but random changes in the game.  The precise split second decisions that led to the ball being in exactly the right place to meet the bat in exactly the right place for a home run rather than a base hit in the first inning will change. It is just as likely that the conditions will change to produce a home run as to prevent one -- it's random. However, as these random changes happen, they cumulate. Even one change such as base hit versus home run may affect how many are on the bases, how many batters come up etc.

    What changes? Why would they affect the game? Why or how are they random? How are they cumulative? Why would these unspecified changes be cumulative? If they are random, maybe they cancel out? Maybe there is no meaningful effect? Are there split second decisions? There is a decision about what to pitch--but it is not split second.  Yes, the batter has to decide when and if to swing in a split second--but why would the dog affect this? Why is it a dichotomy, i.e., positive or negative and not a trichotomy? Why would such a trivial event have any effect on the performance of professional athletes? I have never heard anyone claim that the famous clothing malfunction at the Super Bowl affected the outcome of the game or any player's performance. We are still talking about it. Yet, no one has speculated that it in any meaningful way affected the game. It is hard to write the following without seeming condescending: The reason we do research is to determine if something has an effect or not.

    The reference to superstition was specifically regarding the belief that attending or not attending a sporting event affects the game, not to what you wrote. Have you considered that there may be a reason why they seem condescending? First, you approach me and started this exchange. And, your last posting was prompted by someone liking my posting. And, you have seem to have forgotten the original question in your postings. Yes, I admit the fruit metaphor might have been inappropriate. Yes, I probably was condescending about sports, but someone else made a similar point which you ignored and continue to ignore. Basically, I am reacting to you just throwing ideas out there without evidence or being specific. Pointing out the lack of evidence would seem condescending. One participant before our exchange sent me a private email and ask me if I thought people were serious or if they were joking about how attending a game would affect the home run. 

    It would have been more productive if you suggested another topic. 

    Best wishes.

    Michael





  • 22.  RE: The Nature of Reality

    Posted 01-21-2018 10:33

    You are right Ed.


    I deliberately changed the question to one of statistical modeling. I wanted to ask if there was ever a scenario where the presence of just one more spectator could influence the hitting of a home run that did not involve very inventive counterfactuals.


    The most famous case of one fan actually influencing whether a home run was hit or not hit occurred in the 1996 ALCS when a young boy reached over the wall and caught the ball that in all likelihood would have been caught by the outfielder. For those not familiar here is a link retelling the story.


    https://www.newsday.com/sports/baseball/yankees/16-years-after-the-catch-jeffrey-maier-lives-a-normal-life-1.4098718


    Ed, your discussion and questions have clearly hit some kind of nerve here. I can see why your speculations might be uninteresting to some people on the ASA site but it is difficult to understand the tone of some of the responses.






  • 23.  RE: The Nature of Reality

    Posted 01-21-2018 10:48

    Hi, Michael E:


    (Two Michaels' now commenting on this!):


    I think I find the random counterfactual scenario more interesting. But it doesn't need to involve anything inventive. The point is that it is conceivable that such a small change would produce tiny split second changes that would grow to affect the outcome (but in a random way). There is no specific unlikely counterfactual involved, anymore than Lorentz had one.


    I think also that my recent response to Michael M citing Lorentz's work will probably be my final contribution to this discussion.  Either that makes sense to people on the list (including Michael M) or it does not.


    Thanks for your last line of support about tone. I think my own comments have been consistently respectful.


    Ed



    Ed J. Gracely, PhD
    Associate Professor
    Family, Community, & Preventive Medicine

    College of Medicine

    Associate Professor

    Epidemiology and Biostatistics

    Dornsife School of Public Health

    Drexel University
    2900 W. Queen Lane,
    Philadelphia PA, 19129

    Tel: 215.991.8466 
    | Fax: 215.843.6028
    Cell: 609.707.6965

    eg26@drexel.edu (egracely@drexelmed.edu forwards)
    drexelmed.edu  |  drexel.edu/publichealth






  • 24.  RE: The Nature of Reality

    Posted 01-21-2018 10:41

    Hi, Michael M:


    I changed the scenario from someone in the stands to a dog because even I had a hard time seeing how someone in the stands could make a difference.  I don't think that that change affects the basic "nature of reality" question, which is whether small differences have effects on later outcomes.


    Again, going back to comparing apples to oranges, I wish to emphasize that I am ONLY defending what I called the "orange" scenario of random cumulative effects. Your words still hint at the apple one, which incorporated some kind of predictable effect. 


    I'm sure you've seen it before, but my starting point is Lorentz's weather experiments in which rounding a figure from 6 digits to 3 did NOT cancel out or average out -- it drastically affected the weather predictions over the next few months in his otherwise deterministic models. https://www.technologyreview.com/s/422809/when-the-butterfly-effect-took-flight/


    You ask for evidence.  Since I am not suggesting any systematic effect, what is there to test?  We can't rerun a baseball game with every detail exactly the same but with some tiny event at the start added.  This is closer to philosophy than science.  We can do what Lorentz did and see how complex processes in a deterministic model are affected by tiny changes -- it's been done. The changes cumulate and lead to huge differences. 

    To me it makes sense that the dog walking across the field would function like the rounding of digits by a tiny amount. It doesn't have any systematic effect, but does have a random cumulative effect.  I guess if that doesn't make sense to you, we have to agree to disagree. But the idea is built in to chaos theory and other mathematical models, so it isn't exactly Ed Gracely's invention.

    Ed


    Ed J. Gracely, PhD
    Associate Professor
    Family, Community, & Preventive Medicine

    College of Medicine

    Associate Professor

    Epidemiology and Biostatistics

    Dornsife School of Public Health

    Drexel University
    2900 W. Queen Lane,
    Philadelphia PA, 19129

    Tel: 215.991.8466 
    | Fax: 215.843.6028
    Cell: 609.707.6965

    eg26@drexel.edu (egracely@drexelmed.edu forwards)
    drexelmed.edu  |  drexel.edu/publichealth






  • 25.  RE: The Nature of Reality

    Posted 01-21-2018 11:21
    Ed,

    The point of the experiment was that rounding the number result in a big difference, not that small changes resulted in big differences. Precision was the issue. It is well know that some small changes will result in a big difference in the long run. 

    You apparently do not read the posts. No one is arguing that some small changes do not have a large effect in the long run. That was the point of the short story. But, you did not describe the long run--but a baseball game. And, not every small event has an impact an impact in the long run. Nor, is every event cumulative. Also, you pick an unlikely scenario--that it affects the players.  A delay in game could result in different weather which could affect the game--especially a home run. 

    Of course you can come up with a scenario where there would be a change that affected the game. But, that is not relevant to the original question. And, it is not exactly scholarship to throw scenarios and talk about unknown effects and to assert that they are cumulative, random, and that they must influence the events in some way. 

    Michael







  • 26.  RE: The Nature of Reality

    Posted 01-22-2018 09:46
    Michael M:
    I think Ed is making an important point about the nature of reality - how we know effects - although I would tend to describe the phenomena he is calling "random" (as we in stat are trained to do) as "chaotic", preferring to reserve "randomness" for our information state - whether that is actual as in Bayesian theory, or limiting potential as in frequentist theory - which ranges from zero (completely random) for certain quantum measurements, to practically infinite (so large as to be effectively certain) in classically deterministic settings.
    You said "not every small event has an impact in the long run. Nor, is every event cumulative. Also, you pick an unlikely scenario-"
    I'm curious how you know these things...or rather, how you know the frequency with which "small" events have an impact in the long run, and how likely scenarios are.
    It seems to me events almost always have some causal impact, although that ranges over a distribution which might go from quantum small to quite large (like not getting the home run).






  • 27.  RE: The Nature of Reality

    Posted 01-23-2018 19:59
    The original question was not if small events can produce big changes in the long run. No one disputed that. The question was would reality change in a specific given way if somebody attend or not attend a sporting event. In the first case did it affect the home run.  In the second case, in real life, did it affect who won the game. If the Robin Williams's character had attend the ball game he probably would not have met his wife, at least not at that time and place. But, would he as one anonymous fan affected the home run? Of course, if he shot the batter who hit the home run before the batter came to the plate that would prevent that home run. But, this would involve more than him simply attending the game.
     
    I do not believe that any serious philosopher, mathematician, or scientist has proposed that every small event would have an impact in the long run. (My philosopher collaborator could not think of anyone and neither could I find anyone by doing a quick search.) Just think what that would mean? Here is an example: if I scratched my nose with my left hand just now instead of using my right hand which I normally do there would be long term consequences because of the change. In the Bradbury (sorry if I misspelled the name) story the guides plan their actions, so they are dead ended--that is no consequences. They only killed dinosaurs just before they were going to have a natural death so there would be no consequences. It was one hunter who committed a trivial act (stepping on a butterfly) that changed history. There is a literature on counterfactuals. It is a serious area of discussion. We do computer models to see if small differences matter. Economists analyze what impact small differences make. Not every small action has an impact, some dead end. Some would be canceled out by other events. And, not every small action would influence a specific event
     
    Ed proposed that the dog running onto the field would produce a difference in a baseball game. In his first attempt he gave a casual scenario that was improbable. He then asserted that this event would cause small unknown changes. If they are unknown then you cannot say that they are cumulative, random or have any consequences for the game. Can you know anything about an unknown unknown (this time doubling was on purpose unlike my previous post)? They could be random, and they might not. If they are random would they cancel out? Would they be cumulative? Some might be cumulative? Also, he ignored the possibility that there was no effect on the home run. I came up with a scenario that affected the home run--change in the weather because of the delay, specifically the wind. I asked 2 guys in a sandwich shop how it would change the game. One said it could affect the pitcher's arm--if it was not warm enough. Here you have a causal chain--not unknown unknowns. Here is what was proposed by Ed: many little changes occur, you do not know what they; what process or processes created them; what processes they affect, but you can make definitive statements about them. The statements can only be legitimate if you believe that all small changes are random, cumulative, and have an effect. This would be difficult to justify
     
    Given Ed's title I would assume he knows what random means and that he meant what he wrote.  But since they are unknown unknowns can we say that they are chaotic? I assume by chaotic you meant in the common usage, not the mathematical concept. Here is a discussion of the term in the mathematical sense: http://mathworld.wolfram.com/Chaos.html  Since you suggested that he may mean chaotic and not random I can infer that you do not think they are random. But since we know nothing about them we cannot make statements about them. There could be any underlying pattern. So why would you ask me how I know that they are not random--if you do not think they are random or believe he meant that term? This suggest that you do not think that they are random.
     
    To make clear, I was making the point that he could not make statements about unknown unknowns that are valid. In an early post I made that clear. 
     
    Since I might be wrong please let me know If there is a serious theory that proposes that every trivial action has long term consequences. I would appreciate the reference. If there is some statistical or epistemological technique for determining if unknown unknowns are random or chaotic or cumulative please let me know about it.  
    I also thought the exercise of coming up with scenarios that change the game was a waste of time. It is obvious that you can come up with a scenario that would change a ball game. That is something you can do with friends over a beer. I was opening for a serious discussion since that was the tone of the original posting. The author of the original post mentioned that all his friends and colleagues thought he was wrong and only a 14-year-old boy thought he was right. I assume he was looking for a serious answer. And, I treated his question seriously.

    "I know you won't believe me, but the highest form of Human Excellence is to question oneself and others." 
    ― 
    Socrates
     
    I hope I do the first part. And, I hope I caught all my typos and errors. My last post was an embarrassment. If I did not, please forgive me.
     
    Michael
     
     




     





  • 28.  RE: The Nature of Reality

    Posted 01-24-2018 08:57

    While the conversation is interesting and thought provoking, all of the examples and counterfactuals rely on one thing:  that attending or not attending the game would adversely affect the home run to result in no home run and that the hit would not have resulted in the run scored.  Attending the game could have resulted in a longer home run hit, a shorter home run hit, a fly to center field that was caught, a fly to center field that was missed, a fly to left field, and so on ad infinitum. So, given that changing the reality would have an effect, short or long term, there are an infinite number of possibilities, quite literally.  Therefore, the probability that attending the game would have directly changed only one of them or even a group of them, is 0.  It is exactly the same as the probability of a single exact number occurring in a probability distribution function – 0.






  • 29.  RE: The Nature of Reality

    Posted 01-26-2018 16:09
    Williams's character's absence could have prevented a run by the winning team or many other potential events. That is, had he gone to the game, the winning team could have won by even more, even if the home run everyone is focusing on was unaffected.

    I think the whole thought exercise is some what point less. Wasn't Car Sagan describing the butterfly effect in the 70s. Is this concept really going to be settled here and now?

    So what if William's character would not have met his wife? Maybe he would have met another equally terrific wife at the game. Then the story would be about how he was so glad he did go to the game etc. I think the point of the butterfly effect is that outcomes can (though they don't necessarily) change drastically. Maybe Williams wouldn't have become a psychologist and thus could never end up as a central character in the movie. Maybe his not going to the game would have somehow influenced some chain of events that even would have prevented the birth of Matt Damon's character such that the main character couldn't have existed. Maybe this, maybe that. Maybe had the story about meeting the wife been different the movie would not have been a success and noone would have posted about this on this forum, etc.






  • 30.  RE: The Nature of Reality

    Posted 01-21-2018 01:11

    When I was an adolescent - and a cigarette smoker - and waiting for a bus to take me home from school, we joked that, if we lit up a cigarette, the bus was sure to come.  But it was a joke, even among us adolescents.






  • 31.  RE: The Nature of Reality

    Posted 01-24-2018 05:11
    'You asked, Laurence, for our opinions on your view about the nature of reality.  I have followed the lengthy discussion with interest and have  enjoyed the many intelligent comments even when contradictory.  In light of the more recent comments, perhaps I can contribute another view or at least say some things in a different way.

    First, I appreciate the cleverness of a counterfactual, or fictional, example containing a counterfactual.

    Next, I've been reading a bit about Kant's philosophy, mostly in the online Stanford Encyclopedia of Philosophy.  He aimed to resolve difficulties raised by the determinism of Newtonian physics.  His approach was in part to distinguish between things as we construct them from experience and "things-in-themselves".  I'd rather not get into your view in terms of the latter because of the issues involved.

    After reading a few opinions of others, I think I understood your view better, which I shall consider in terms of knowledge from experience.  You think all things are flukes in that they occur precisely as they do because their immediate causes were precisely as they were.  Any change in conditions and what would happen would be a fluke that's at least a little different. There is some truth in it.  Carlton Fisk's name is vaguely familiar to me, but I'm not a sports enthusiast, like the one Michael M. met in a sandwich shop, nor a sports statistician.  But I take it Fisk really hit a famous home run in the World Series, and you may consider it to have required a particular precise alignment of conditions, and "The Play" similarly.

    I think it's pretty obvious according to the knowledge human understanding has constructed from experiences that, for prediction of some phenomena, there are things that are of no practical help.  And some of these phenomena can be predicted with great accuracy.  Solar eclipses come to mind.  In this body of empirical knowledge, they are not flukes.  And your view of the nature of reality, that everything is a fluke, is falsified.

    Of course, falsehood of this view does not mean that predictions based on it are wrong in your two examples or in a larger subset of all cases.  I take it your two examples are well known to many people in part because they would not have been predicted even by baseball and football experts, based on experiences.  And I think they similarly would not have been predicted under other circumstances.  You're right insofar as the plays probably wouldn't have happened if particular people had attended.  But, until the plays did happen, they were unlikely.  The absence of particular people was not predictive, nor causative.

    I've heard there really isn't a home-field advantage.  But, if crowd size would make a difference, which Michael M. disputes, I see a way someone's attendance could affect the predictive probability of a particular event in a game.  It's simply that the probability a randomly selected person attends a game is proportional to crowd size.

    I think current empirical knowledge gives a different view of the nature of reality from the view you expressed.

    ------------------------------
    Thomas Davis
    ------------------------------