NOW LIVE! Today's the day you meet your new best friend. You don’t have to leave Wolfy behind... In 'Pets & Sidekicks' your companions level up with you!

Reply to thread

Message: <blockquote data-quote="clearstream" data-source="post: 8626592" data-attributes="member: 71699">Characterising my comment as a non-sequitur makes me suspect we are speaking past one another. We're likely equally mystified by the other's position!Just for avoidance of doubt, I am not suggesting that the specific words capturing a theory must include the word "predict" or its variants in order to be predictive.If A contributes to B such that in the presence of A, B is more likely, then it is correct to say that A predicts (with some probability) B. There might be a misapprehension that when I use "predicts" I mean with certainty. To ensure that's not getting in the way, I mean "predicts at some probability, itself unknown". Hypothetically, factor-based analysis might be done to establish if mixtures of agendas are predictive of dysfunction using Bayesian algorithms. That'd be a tremendous amount of work, and not something we're likely to have the resources to embark on!Weakness or scattered manfestation of consequence isn't of itself a failure to count as prediction. This suggests to me again that we may have different expectations as to "predicts". A predicts B at some probability, which need not be 1. We can say that in the presence of A, B is more (or less) probable, and talk about the power and significance of any such correlation. I believe that is what Edwards is claiming: incompatible agendas "contribute". The qualification isn't a taking back of "contribute", it's saying that the nature of dysfunctional play is not tautologically the simple noticing of incompatible agendas (i.e. realising one is in the presence of agendas that are incompatible), dysfunction is its own phenomena. Love is perhaps one of the factors that may contribute to childbirth, so that childbirth is more likely in the presence of love, but love isn't childbirth. Maybe others can find better examples?As I posted (to [USER=22779]@Hussar[/USER] and [USER=6790260]@EzekielRaiden[/USER]) perhaps the key misgiving is that to say predicts sounds like entailing with certainty, whereas your position might be that the presence of incompatible agendas (once identified) is only at some low degree of signficance correlated with the presence of dysfunction. Note this isn't a claim that the former causes the latter, although one might want to investigate the possibility of a casual relationship. Even were a causal relationship established, it need not ignore complexity: incompatible agendas could (hypothetically) cause dysfunction at some degree of likelihood or only in the presence of some additional factor, C. We need not assume the factors are sufficient and invariant, or that the values are static.I feel the analogy of an auto-repair manual does not serve your argument (as I understand it) very successfully, as one does predict that when one follows the manual one will achieve the specified result. A good manual is a case of strong and confident predictability.For example, I recently noticed a new warning light on my VW dashboard. Following the manual I saw that it warned of a lighting fault. Based on that, I made the prediction that one of the lights was out and checked them all. The prediction turned out right. I turned to the manual for the size and type of lamp to buy. Based on the manual I felt optimistic in predicting that a given lamp would match the fitting. It did. Suppose however that I had ignored the manual and chosen a lamp at random: could I have predicted that it would fit with any confidence?I'm especially sympathetic to the point that human behaviours are extremely complex. I mean, think of generalised intelligence which has been subject to a tremendous amount of factor-based analysis as to what it predicts! I additionally believe GNS theory is unrepresentative in important respects, noted by others. Should I expect an unrepresentative theory to predict an exceedingly complex phenomena in a testable way?No, in the end I agree with you. Not as to prediction, but let's set that aside. GNS has important limitations to bear in mind going forward. There's a high risk of refitting the facts to the theory. Emphasising the parts that seem to fit. We'll find I think that multiple theories will explain what we observe equally well, and we cannot exclude ourselves from our analysis. One person might apply GEN two-tier theory and see the features fall into place, everything becoming less puzzling, and arrive at new and helpful insights. Another preferring or getting more out of GNS. That might be excluded by testing and falsifying, but as we are agreeing that we cannot test and falsify we cannot exclude it. And all the theories are likely to be improvable.</blockquote>

Verification