------------------------------------------------------------------------ Subject: POSITION PAPER #1 CRITIQUE ------------------------------------------------------------------------ Now that you've gotten your graded and revised PP#1 back, I thought that some of you might find it useful to see how I might critique a position paper. At the same time, this will give you an idea of what I mean by "active reading" of the sort that you should be doing in your Reading Journals. What follows is an active-reading critique of one of your position papers, chosen randomly (and anonymously!). ------------------------------------------------------------------------- My comments are indented and set off by single lines, like this sentence. ------------------------------------------------------------------------- ========================================================================= Position Paper 1 What is Computer Science? Given to us was: 1: Science is the systematic observation, description, experimental investigation, and theoretical explanation of natural phenomena. 2: Computer science is the study of computers and related phenomena. 3: Therefore, computer science is not a science. Position: My Position on this matter is that I don't exactly agree with the first ground, that science is the systematic observation of natural phenomena, but a phenomenon rather. ------------------------------------------------------------------------- I'm not sure I understand this; perhaps the problem is one of grammar. Do you mean that you think that science is a phenomenon itself? Or do you mean that you think that science is the systematic observation (and description, etc.--you left all that out, but I assume you are including it) of "a phenomenon" rather than of "natural phenomena"? The first interpretation may be a true statement, but doesn't seem to be relevant to the issue. The second interpretation is still puzzling: Which single phenomenon is science a study of? Perhaps you mean something else altogether, such as: Science is the systematic observation (etc.) of *any* phenomenon, natural or otherwise? That would make more sense. But if that's what you mean, you'll need to explain why you think that science has that kind of broad subject matter. Maybe you'll say something more about it later... ------------------------------------------------------------------------- The second position given I agree with. So ground 2 holds true. Ground 3 on the other hand I disagree with. ------------------------------------------------------------------------- Perhaps this is just a poor choice of words, but statement 3 is not a "ground"; it's a conclusion. A "ground" is a reason for believing something else. So, the premises of an argument, like statements 1 and 2 above, are grounds for the conclusion (statement 3). Since you already have said that you disagree with premise 1, that could explain why you disagree with the conclusion. Let's see... Also, I note that you do not say *why* you agree with the second "position" (I assume you mean "premise"). ------------------------------------------------------------------------- I do agree that computer science is indeed a science because science is defined as the observation, description, experimental investigation, and theoretical explanation of phenomena. ------------------------------------------------------------------------- A small problem with wording: You can't "agree" with this statement, since it's not part of the argument. You *believe* it, but there's nothing to "agree" with. What you're presenting here is a new argument: 1'. Science is the (systematic??) observation, description, experimental investigation, and theoretical explanation of phenomena. 2. (as above) 3'. Therefore, computer science *is* a science. Now, you still haven't said why you disagree with 1; you've merely offered a *replacement* for it. You need to give some justification of why science studies all phenomena and not just "natural" phenomena. Also, both arguments have a missing premise! The first argument assumes that "computers and related phenomena" are *not* "natural" phenomena. Your replacement argument assumes that "computers and related phenomena" *are* "phenomena". You need to say why you think the first statement might be false and why you think the second statement might be true. ------------------------------------------------------------------------- Science is also defined as an activity regarded as requiring study and method(1). [Footnote (1): The American Heritage "dictionary" 2001]. ------------------------------------------------------------------------- Perhaps, but how is this relevant? You agree with premise 1 that science is the (systematic??) observation, etc.; you merely disagree on what it is the (systematic) observation, etc., *of*. Adding this *new* definition seems to be irrelevant. Also, for what it's worth, the correct name of the book is The American Heritage Dictionary, without quotes around "dictionary". ------------------------------------------------------------------------- Natural science however is defined as a science, such as biology, chemistry or physics that deals with the objects, phenomenon, or laws of nature and the natural world.(2) [Footnote (2): [same as above]] ------------------------------------------------------------------------- This certainly seems to agree with premise 1, although it does qualify "science" as "natural". So maybe your point is that science (with no qualifiers) studies phenomena (with no qualifiers), whereas *natural* science studies *natural* phenomena. If so, it would be nice to have an example of a non- natural science (besides computer science). By the way, you need commas around "however", a comma after "physics", and "phenomenon" should be "phenomena" (plural). ------------------------------------------------------------------------- I am not defining Computer science to be a science of the natural world, rather a science, which is involved in observation, description, experimental investigation, meaning algorithms, and theoretical explanation of phenomena. ------------------------------------------------------------------------- This just repeats what you've said before, still without giving any reasons for believing it. But there is one new thing here: "meaning algorithms". However, I don't understand exactly what you mean. Are you talking about some special kind of algorithm called a "meaning algorithm" (maybe something used in natural- language processing???); probably not. Maybe you are explaining what "experimental investigation" is; are you saying that "algorithm" is a synonym for "experimental investigation"? But there are several problems with that: First, this definition of science is meant to be understood as follows: Science is the observation of (natural) phenomena, the description of (natural) phenomena, the experimental investigation of (natural) phenomena, and the theoretical explanation of (natural) phenomena. But it makes no sense to say "the algorithms of (natural) phenomena". So "algorithm" can't mean exactly the same thing as "experimental investigation". Second, we know what algorithms are: They are computer programs, procedures, things like recipes or the rules for doing long division. But those kinds of things are not experimental investigations, or at least not obviously so. ------------------------------------------------------------------------- Algorithms is a precisely defined sequence of rules telling how to produce specified output information from given information from given information in a finite number of steps.(3) [Footnote (3): Knuth "CS and Its Relation to Math" 1974]. ------------------------------------------------------------------------- The first word should be "An algorithm", and you've got too many "from given information"s :-). OK--so we have Knuth's definition. Now what?... ------------------------------------------------------------------------- Since algorithms and formulas can be used in physics and chemistry and be considered a science, then so can computer science. ------------------------------------------------------------------------- Ah! Another argument! This one looks like this: A. Algorithms [and formulas, but let's ignore them] can be used in physics and chemistry [which, recall, are natural sciences]. B. [Therefore???] Algorithms can be considered a science. C. Therefore, computer science can be considered a science. The first premise seems OK. What about the second premise? Earlier, you defined science as the systematic observation (etc.) of phenomena (natural or otherwise). I'm still not happy with that, but let's accept it for now. So premise B seems to mean that algorithms are the systematic observation (etc.) of phenomena. But that's not what an algorithm is. So premise B seems false. Maybe you think that it follows from premise A; it's not clear from what you wrote. But the only way it could follow from premise A is if you assume some general principle such as this: Anything that can be used in a (natural) science can itself be considered a science. (Then, since algorithms can be used in science, they must be a science.) But I can use microscopes, test tubes, pencils, pens, paper--all sorts of things--in science, and no one would consider *them* to be sciences. So, that general principle seems false, and premise B doesn't seem to follow from premise A. Suppose for the sake of argument that we accept both A and B. Does C follow? Actually, C might follow from B alone, but only if we assume something like "algorithms are part of computer science" and "if something has a part that's a science, then it's a science, too". How about this possible counterexample: Suppose that B is true. Now, algorithms are used in astrology (you need algorithms to figure out the positions of the heavenly bodies). Does it follow that astrology is a science? I don't think so, but maybe you do? ------------------------------------------------------------------------- [end of essay] OK: Note that I haven't summarized things, or given my own views on the subject. I've merely raised some puzzles, concerns, questions, possible counterexamples, etc., that (I hope!) can help the author of this position paper clarify his or her position, and maybe write a better second draft. Or, to use a computational metaphor: If you consider this draft to be a "program" designed to "manipulate" my beliefs, then it has a few "bugs" in it (since I didn't agree with it). So, it needs to be "debugged" :-)