tag:blogger.com,1999:blog-15005476.post6626999654958402986..comments2023-10-10T08:02:18.073-04:00Comments on Rationally Speaking: Toward a science of morality. An annotated response to Michael Shermer.Unknownnoreply@blogger.comBlogger116125tag:blogger.com,1999:blog-15005476.post-84509182005049649302013-03-02T09:09:59.602-05:002013-03-02T09:09:59.602-05:00Hector, isn't your list of 3 an attempt to bri...Hector, isn't your list of 3 an attempt to bridge the is-ought divide? I.e, 1 = 'is'; 2 = 'ought'; and 3 is what Shermer (and Sam Harris) are trying to argue for. In my view, their arguments are wanting, precisely because they believe they can foist "certain assumptions and values" onto a moral system, as if such things exist as objectively true concepts.Philboid Studgehttps://www.blogger.com/profile/12303760443013349885noreply@blogger.comtag:blogger.com,1999:blog-15005476.post-91557295839713304322013-03-02T09:09:21.143-05:002013-03-02T09:09:21.143-05:00Hector, isn't your list of 3 an attempt to bri...Hector, isn't your list of 3 an attempt to bridge the is-ought divide? I.e, 1 = 'is'; 2 = 'ought'; and 3 is what Shermer (and Sam Harris) are trying to argue for. In my view, their arguments are wanting, precisely because they believe they can foist "certain assumptions and values" onto a moral system, as if such things exist as objectively true concepts.Philboid Studgehttps://www.blogger.com/profile/12303760443013349885noreply@blogger.comtag:blogger.com,1999:blog-15005476.post-51657183870289975652013-02-28T09:46:55.797-05:002013-02-28T09:46:55.797-05:00If you'd prefer that I didn't answer then ...If you'd prefer that I didn't answer then we can drop the conversation.Disagreeable Mehttps://www.blogger.com/profile/15258557849869963650noreply@blogger.comtag:blogger.com,1999:blog-15005476.post-87066252639529650792013-02-27T18:36:26.580-05:002013-02-27T18:36:26.580-05:00You don't seem to be able to grasp that we can...You don't seem to be able to grasp that we can both make and use algorithms which then become a part of us, and yet have to be more than algorithms to make the living system work - while a computer can use its man made algorithms but can neither make nor use them without a human's help to program them. <br />In short, humans program and reprogram their algorithms, Computers don't and can't unless the humans help them do it.<br />And it's not at all irrelevant that Kasperov beat the computer because it's evidence that the computer depends on its programmer for its use of that programmers intelligence. Otherwise if the intelligence of the programmer had nothing to do with the computers intelligence, it should have beaten that mere human, Kasperov., on its own.<br />And then you make no attempt to explain why the newer computer can certainly beat any human. Why, because it has had a better programmer, or because it has self evolved its algorithmic system, or what?<br />I'd rather that you didn't answer, but of course you will. You seem to have no sense of the consistency that's necessary for the sake of logic. But then neither do computers if their programmers don't.Baron Phttps://www.blogger.com/profile/04138430918331887648noreply@blogger.comtag:blogger.com,1999:blog-15005476.post-21364581586236687402013-02-27T17:45:12.321-05:002013-02-27T17:45:12.321-05:00>No, they're not, if the purpose of the gam...>No, they're not, if the purpose of the game is to win it. Or else you know next to nothing about playing chess.<<br />I have to say flat out that you're wrong here.<br /><br />Historical data is useful but not absolutely essential. Give a computer enough processing power and memory and any advantage from mining historical data can be overcome through brute force.<br /><br />>And Jesus, where in hell did I say we can't be algorithms?<<br />Apologies. Your precise position is hard for me to grasp. If you agree with me that we are algorithms then we can build on that.<br /><br />However, your subsequent statements that algorithms might need additional "biological assessment processes" or that human brains are capable of doing stuff that computers cannot rather indicates that you do in fact believe that we are more than algorithms.<br /><br />By definition, an algorithm is something that can be carried out by a digital computer. If you think that we are capable of more than computers, then you don't think that we are algorithms.<br /><br />>We can't create machines that compute intuitively, and computers can't simulate our intuitive human brain processes.<<br />Yet. There's no good reason to believe that this is impossible in principle.<br /><br />>But as it turned out in the famous game between Kasparov and the IBM computer in 1996, Kasparov won the game. Because in part, the computer's creator was no Kasparov.<<br />Irrelevant. The current state of the art chess computer is unbeatable by any human, Kasparov included.Disagreeable Mehttps://www.blogger.com/profile/15258557849869963650noreply@blogger.comtag:blogger.com,1999:blog-15005476.post-79449041265287669062013-02-27T15:36:08.732-05:002013-02-27T15:36:08.732-05:00"chess computers are also capable of playing ..."chess computers are also capable of playing chess without using historical data."<br />No, they're not, if the purpose of the game is to win it. Or else you know next to nothing about playing chess.<br /><br />And Jesus, where in hell did I say we can't be algorithms? Our brains must use their own evolved forms of predictive algorithms to function. Which couldn't happen either if our sensory apparatus didn't furnish the environmental observations from which the most viable optional approaches could be induced. <br />What I said of course is that your algorithms can't simulate brains in a computer without simulating our sensory apparatus. Plus without an additional biological assessment process that programs those algorithms differently to suit each organism's particular purposes. (Purposes that computers can't acquire either without our brains.)<br />And computers that are programmed to play chess don't "expect" anything. Expectations are built into the programs as its alternatives. We may be surprised by the results that the complexity of the process produces, but that's why we built it - to use binary systems to solve problems mathematically in a few seconds that would take our human brains a few years to do. But only our human brains have the ability to create mathematical problem solving machines. We can't create machines that compute intuitively, and computers can't simulate our intuitive human brain processes.<br />And in the end of course the chess computer is supposed to make better moves than its creator would. But as it turned out in the famous game between Kasparov and the IBM computer in 1996, Kasparov won the game. Because in part, the computer's creator was no Kasparov.<br />Baron Phttps://www.blogger.com/profile/04138430918331887648noreply@blogger.comtag:blogger.com,1999:blog-15005476.post-49112342458927865822013-02-27T14:00:12.842-05:002013-02-27T14:00:12.842-05:00>The chess computer was made to beat their prog...>The chess computer was made to beat their programmers because it was given all the known options that human chess players were known to have devised in the past.<<br /><br />Historical data was used and it was of benefit for the computer, however chess computers are also capable of playing chess without using historical data. Some of these chess computers can beat their programmers without reference to historical data.<br /><br />Everything else you say in this paragraph about chess computers is true and beside the point.<br /><br />>You seem to think, in other words, that algorithms somehow construct themselves.<<br /><br />No I don't.<br /><br />>If the maker is surprised at what an algorithm can do, it was not the algorithm that surprised him, but the consequences of his own creation that were not anticipated.<<br /><br />The algorithm surprised him. The consequences of his own creation surprised him. Both are true and equivalent.<br /><br />If we get back the point, I believe your argument was that algorithms can't cope with the unexpected because they can only do what they were designed to do. My answer is that whether solutions are found by the algorithm or as consequences of the designer's creation, solutions may yet be found to unexpected situations which are superior to those that might be found by the designer himself.<br /><br />To bring it back to the chess computer, there are positions in chess the designer will never have considered (and so in a sense they are unexpected) but the chess computer will still be capable of dealing with them. If the chess computer is better at chess than its designer (a likely scenario), then the chess computer will make a better move than its designer would in that situation.<br /><br />This refutes your point that we can't be algorithms because we can deal with the unexpected.Disagreeable Mehttps://www.blogger.com/profile/15258557849869963650noreply@blogger.comtag:blogger.com,1999:blog-15005476.post-80826635563454257012013-02-27T13:23:51.562-05:002013-02-27T13:23:51.562-05:00DM,
The chess computer was made to beat their prog...DM,<br />The chess computer was made to beat their programmers because it was given all the known options that human chess players were known to have devised in the past. It was given a memory to use that humans don't have the capacity to memorize. It didn't devise any options from any clues that weren't already entered in the system by its programmers. It didn't select what it either needed or wanted to remember, and it didn't do any "thinking" that the programmers hadn't fashioned it to do. It was nothing more than a very complex tool that couldn't work without its programming coming to it from outside of itself.<br />You seem incapable of getting that. You seem to think, in other words, that algorithms somehow construct themselves. If the maker is surprised at what an algorithm can do, it was not the algorithm that surprised him, but the consequences of his own creation that were not anticipated. The algorithm itself is incapable of either creating a surprise or being surprised itself.<br />But you'll now protest that you didn't understand what I've just written. And that it's my fault. And because I'm incapable of explaining what you're not capable of understanding, you'll be right.Baron Phttps://www.blogger.com/profile/04138430918331887648noreply@blogger.comtag:blogger.com,1999:blog-15005476.post-66825908790319661142013-02-27T08:10:40.300-05:002013-02-27T08:10:40.300-05:00> If a thing decides according to some algorith...> If a thing decides according to some algorithm, did the algorithm derive its options from its own sensory apparatus, as life does, or were these options built in or later added by the human that prepared them for that algorithms use? <<br />Don't know what you mean about options - in the simplified example of a mathematical function I suppose the available options consist of the output space, namely the numbers. I did not prepare them for the algorithms use (in the sense that I did not list out all possible answers in advance), but the options available are implicitly implied by the nature of the algorithm I have designed.<br /><br />The options available to animals are the space of possible actions they can take, which can be further decomposed into what muscle movements they can make, which can be decomposed into what nerve signals they can send. The space of what nervous signals they can send is predetermined by their evolutionary heritage.<br /><br />>Or perhaps you've been the first to come up with algorithms that can choose to use lifelike sensory apparatus to prepare its fellow algorithms.<<br /><br />You've lost me.<br /><br />>And if you can't predict what an algorithm's choice will be, you've made one that makes choices by randomness<<br />Not so. Lots of algorithms are complicated and the only way to figure out what choices they will make is to work out the steps the algorithm will take step by step. For sufficiently complex algorithms, this is not feasible for a human being armed with pen and paper to calculate. (this is why we have computers, after all!).<br /><br />The only way to see what such an algorithm will do is to run the algorithm on a computer, and so it is possible for the algorithm to surprise us. This is why it is possible for chess computers to beat the programmers who made them, despite there being no randomness involved.<br /><br />And so I'm not sure the rest of your argument is relevant since your premises seem to be incorrect.Disagreeable Mehttps://www.blogger.com/profile/15258557849869963650noreply@blogger.comtag:blogger.com,1999:blog-15005476.post-47453245071629028112013-02-26T15:49:05.905-05:002013-02-26T15:49:05.905-05:00Just to follow up, I was a bit unfair. Your commen...Just to follow up, I was a bit unfair. Your comments about the mechanism in the predetermined niche are understandable in the context of my comment to which you were replying to.<br /><br />I see now that the mechanism is something implementing the function "return twice the number" and the predetermined niche is the job I have assigned that function.Disagreeable Mehttps://www.blogger.com/profile/15258557849869963650noreply@blogger.comtag:blogger.com,1999:blog-15005476.post-42073306672449786312013-02-26T15:29:16.200-05:002013-02-26T15:29:16.200-05:00"Rather, it needs me to build it so that it c..."Rather, it needs me to build it so that it can decide for itself according to some algorithm. It may be that the algorithm is so complex that it would be unfeasible for me to predict what its choice might be."<br />Good god, man, that's pitiful. If a thing decides according to some algorithm, did the algorithm derive its options from its own sensory apparatus, as life does, or were these options built in or later added by the human that prepared them for that algorithms use? <br />Or perhaps you've been the first to come up with algorithms that can choose to use lifelike sensory apparatus to prepare its fellow algorithms.<br />And if you can't predict what an algorithm's choice will be, you've made one that makes choices by randomness - and since we've supposedly been talking about choices made with at least a minimum of intelligence involved, then you've continued to believe that randomness can produce intelligence by accident.<br />But on the other hand, if you consider all of this an adequate explanation, then for you it is. <br />Intelligible, comprehensible, understandable, cogent, and coherent, it isn't.<br /><br />Baron Phttps://www.blogger.com/profile/04138430918331887648noreply@blogger.comtag:blogger.com,1999:blog-15005476.post-47593742869585361692013-02-26T14:29:01.233-05:002013-02-26T14:29:01.233-05:00Baron:
I'm sorry you find me to be incoherent...Baron:<br /><br />I'm sorry you find me to be incoherent, but I must also confess that I find it excruciatingly hard to understand you.<br /><br />For example, what do you mean by "indeterminate"? The word means vague or undefined. I suspect you mean non-deterministic. Why do you think I'm putting myself in a non-deterministic universe?<br /><br />What is the mechanism you refer to? Evolution? An evolved organism? I have no idea.<br /><br />Predetermined niche? Who or what has predetermined it? What do you mean by predetermined?<br /><br />But if I can guess at the general point of the question, you're asking me to recognise the difference between me making choices, as an undesigned entity, and a machine making choices as a designed entity.<br /><br />If I understand you correctly, you are saying that designed entities have their choices predetermined in some sense by the designer, while undesigned entities do not have a designer so do not have their choices predetermined.<br /><br />I answer that I am designed in a sense. My designer is the blind unintelligent force of evolution by natural selection influenced by a great deal of chance happenstance. My choices are predetermined by these forces just as are the choices of a machine I design manually.<br /><br />Furthermore, I'd like to point out that a designed machine does not need me to "select, decide and make its choices for it". Rather, it needs me to build it so that it can decide for itself according to some algorithm. It may be that the algorithm is so complex that it would be unfeasible for me to predict what its choice might be. <br /><br />This is why programmers are often surprised by the output of their programs. Usually, this manifests as software bugs, but occasionally some interesting and useful behaviour can arise unexpectedly.Disagreeable Mehttps://www.blogger.com/profile/15258557849869963650noreply@blogger.comtag:blogger.com,1999:blog-15005476.post-81955347278371032562013-02-26T13:18:12.407-05:002013-02-26T13:18:12.407-05:00In your example, it seems you're putting yours...In your example, it seems you're putting yourself in an indeterminate universe dealing with a mechanism in its own predetermined niche. You have found choices you can decide to make, while the machine however needs you to select, decide, and make its choices for it.<br />So it's not so much that I don't understand you, as it's likely that you don't understand yourself.Baron Phttps://www.blogger.com/profile/04138430918331887648noreply@blogger.comtag:blogger.com,1999:blog-15005476.post-3418326929175182012013-02-26T10:03:36.477-05:002013-02-26T10:03:36.477-05:00There is a moral obligation to reduce the sufferin...There is a moral obligation to reduce the suffering of the existing population as well as to treat and remove the cause of the suffering in the next generation.<br /><br />Policies that only seek to feed existing people are treating only the consequences and not the causes of that hunger. The best programs will treat both. <br /><br />Reducing suffering in the present without reducing the causes of the next generation is condemning the next generation to suffer more of the problem that the current program is trying to combat. That is, the care and responsibility that is provided at one level is offset by failure to reduce the source of the problem for the next generation. Such programs are both compassionate and irresponsible.Anonymoushttps://www.blogger.com/profile/00265673638659855406noreply@blogger.comtag:blogger.com,1999:blog-15005476.post-87277220759812334012013-02-26T08:03:13.148-05:002013-02-26T08:03:13.148-05:00"If the system can't account for every po..."If the system can't account for every possible eventuality, then it's not deterministic, it's probablistic."<br /><br />You misunderstand me. I said it does not need to account for every eventuality in advance, this doesn't mean that it doesn't have a well-defined output for every input.<br /><br />For example, suppose I define a function for integers which gives an output which is twice the value of the input. I have not listed out all possible inputs in advance, and yet it has a perfectly defined output for all inputs.<br /><br />So what happens if I give an unexpected input, such as a letter? Well, if the function is a mechanical system such as a computer program or a biological brain it's going to do something. What it does depends on how it is implemented. It might return a special value such as "undefined" or it might return 0. The point is that even if you give it an unexpected input, it will still have a defined and (in principle) predictable answer.Disagreeable Mehttps://www.blogger.com/profile/15258557849869963650noreply@blogger.comtag:blogger.com,1999:blog-15005476.post-20503951031533262412013-02-25T17:44:38.394-05:002013-02-25T17:44:38.394-05:00DM says, "It is a dynamic process that maps i...DM says, "It is a dynamic process that maps inputs to outputs in a deterministic but complex fashion. It doesn't have to account for every possible eventuality in advance."<br />If the system can't account for every possible eventuality, then it's not deterministic, it's probablistic. And since those systems are indeterminate systems, they will require choices made by any organisms that need to predict and separate the most probable from the many possibles in order to continue to survive. This requires the ability to acquire and apply knowledge and skills, which is the definition of intelligence. <br />Rocks don't have that ability, and neither does water. There are other entities besides life that may have it, but there should be no doubt in the minds of the logical that life has that ability.Baron Phttps://www.blogger.com/profile/04138430918331887648noreply@blogger.comtag:blogger.com,1999:blog-15005476.post-61676366526573798622013-02-25T17:10:03.457-05:002013-02-25T17:10:03.457-05:00If I understand your question, you are asking how ...If I understand your question, you are asking how life forms deal with the unexpected when everything they do is predetermined by their programming.<br /><br />This is because their programming is not a script. It is a dynamic process that maps inputs to outputs in a deterministic but complex fashion. It doesn't have to account for every possible eventuality in advance.<br /><br />In many cases, the programming the life forms have evolved is sufficient for day to day life and can even be adapted to deal with novel problems, as exemplified by humans.<br /><br />This programming is often found to be unreliable in situations which the life form has not evolved to cope with, and so will likely not work particularly well in unexpected situations.<br /><br />We can see examples of this every day, like when birds fly into glass windows. The birds' programming, which evolved in an environment without glass, does not deal with the unexpected glass in an optimal fashion.<br /><br />But it's not just animals that have this problem. The growing problem of obesity demonstrates how the programming of many humans (including yours truly!) does not deal very well with the unexpected widespread availability of more calories than we need.Disagreeable Mehttps://www.blogger.com/profile/15258557849869963650noreply@blogger.comtag:blogger.com,1999:blog-15005476.post-88426777811556177982013-02-25T13:02:10.052-05:002013-02-25T13:02:10.052-05:00DM,
Has it then happened that life forms are prog...DM, <br />Has it then happened that life forms are programmed to have expectations, but no choice, as you see it, in dealing with the unexpected? Since as you see it, programmed expectations aren't expected to be other than anticipated?Baron Phttps://www.blogger.com/profile/04138430918331887648noreply@blogger.comtag:blogger.com,1999:blog-15005476.post-4888504932305065252013-02-25T06:29:23.890-05:002013-02-25T06:29:23.890-05:00> I'd think that such determination would b...> I'd think that such determination would be hard to have arranged for by an eternally endless series of obedient accidents.<<br /><br />You can like to think what you please, but that's not really an argument. There's nothing obedient about the accidents - there is nothing to obey. If your point is that it seems improbable that life would be created by a series of chance events, then you may be right.<br /><br />But the universe is vast, and perhaps infinite. However unlikely something is, in all that vastness you can expect that it will happen, many times over.Disagreeable Mehttps://www.blogger.com/profile/15258557849869963650noreply@blogger.comtag:blogger.com,1999:blog-15005476.post-32090504548439925502013-02-24T22:15:40.522-05:002013-02-24T22:15:40.522-05:00Sorry but you've got yourself a determinant un...Sorry but you've got yourself a determinant universe where no-one has other than an illusion of choice, that effectively perpetuates that illusion for all conceivable time, and there's no determiner? I'd think that such determination would be hard to have arranged for by an eternally endless series of obedient accidents.Baron Phttps://www.blogger.com/profile/04138430918331887648noreply@blogger.comtag:blogger.com,1999:blog-15005476.post-74732128426970211122013-02-24T18:47:29.520-05:002013-02-24T18:47:29.520-05:00Except that it isn't.
There's not an ulti...Except that it isn't.<br /><br />There's not an ultimate programmer, there's an ultimate program: the laws of nature.<br /><br />Now, where do these laws come from? One answer might be that they were created by a programmer, sure. However that's not much of an explanation, because then you have to wonder where the programmer came from.<br /><br />There are other explanations. For example, there might be an infinite number of universes, one for every possible set of laws of nature. Why not?<br /><br />Whatever the actual reason, and I have my own suspicions in this regard, there is not enough to justify a belief in a "final programmer". And even if you do find yourself believing in this programmer, there's no reason to believe that he intervenes, or that he wants us to be good, or that any specific religion is accurate.Disagreeable Mehttps://www.blogger.com/profile/15258557849869963650noreply@blogger.comtag:blogger.com,1999:blog-15005476.post-80779010148533288582013-02-24T16:37:00.496-05:002013-02-24T16:37:00.496-05:00You forgot to point out that unless there's wh...You forgot to point out that unless there's what could be called a final programmer, it's programs programming programs all the way down.Baron Phttps://www.blogger.com/profile/04138430918331887648noreply@blogger.comtag:blogger.com,1999:blog-15005476.post-184203679005930782013-02-24T15:43:58.031-05:002013-02-24T15:43:58.031-05:00I think that probably sums it up.
However, I don&...I think that probably sums it up.<br /><br />However, I don't believe in non-deterministic free will, so ultimately I think we are just as programmed as the rock. Choice is really an illusion.<br /><br />But it's a useful illusion or abstraction when discussing intelligent agents. It's just not fundamentally distinct from the deterministic outcomes of unintelligent events.Disagreeable Mehttps://www.blogger.com/profile/15258557849869963650noreply@blogger.comtag:blogger.com,1999:blog-15005476.post-52405191772886817872013-02-24T14:00:40.358-05:002013-02-24T14:00:40.358-05:00Largely agree with you here, especially on his exa...Largely agree with you here, especially on his exaggerated claims for science, apart from the fact that I enjoy reading Harris's controversial positions nonetheless.Disagreeable Mehttps://www.blogger.com/profile/15258557849869963650noreply@blogger.comtag:blogger.com,1999:blog-15005476.post-33829044999430669222013-02-24T13:06:34.374-05:002013-02-24T13:06:34.374-05:00So early life forms didn't make choices, they ...So early life forms didn't make choices, they were programmed. (Like falling rocks are programmed to kill what they hit, and water is programmed to run downhill and drown stuff.) <br />And intelligence evolved as a talent, like the musicality that life forms also evolved from no initial talent. Neither of which was needed when they operated from a non-evolved program. <br />Nobody never 'splained it like that before.Baron Phttps://www.blogger.com/profile/04138430918331887648noreply@blogger.com