I read a couple of good books recently, and wanted to share them and do some writing to collect my thoughts on a subject that is currently of news-worthy relevance and of particular interest to "Soylentils". Enjoy, and I look forward to the discussion!
(Score: 2) by AthanasiusKircher on Monday May 08 2017, @04:04PM (6 children)
I'll thank to contributor for the review, which was an interesting and thoughtful read.
Though to me there seems to be an unnecessary and counterproductive conflation of science (or "scientific inquiry") with technological shifts. The critiques in the books, as I understand them (I looked at Technopoly a bit years ago, but I haven't read the other, though I've heard of it) are mostly about technology. But talking about "progress" in technology strikes me as very different from talking about "progress" in science. Certainly technological advances can follow scientific ones and vice versa, but science is generally about trying to find an increasing "better fit" to explanations of the workings of the natural world. Technological "progress" has no such metric.
One can say that Einstein's theory of general relativity is a better model of physics than Newton's theory of mechanics in a somewhat objective fashion. (I'm setting aside deeper philosophical issues about empiricism for the moment.) One cannot generally make similar statements that a self-propelled power mower is objectively "better" than an old-fashioned power mower, or that both are objectively "better" than a non-powered reel mower. The former are seen to be technological "advances," but that judgment is based on some sort of subjective utility measure based on various cultural assumptions. If I really dislike noise or I'm concerned about pollution and use of fossil fuels or I just like the way it cuts my grass, I might think the reel mower is a better technological option. Whereas I can't really say that in Einstein vs. Newton -- clearly it's easier to get an estimate at normal speeds, etc. with Newton's theory (which is why we still teach it as a tool in intro courses), but there's never any question about which is actually a more accurate or complete theory.
This distinction is crucial because the review starts with some meditations on whether "science 'serves the common good', and that the process of scientific inquiry is the best path to finding solutions to cultural and and societal problems" only to spend most of the review on technological "progress."
And then there's the notion of "progress" in general, which is insufficiently interrogated. The problem with "cultural and societal problems" is that their "solutions" and even their very definitions are also cultural and societal, which introduces even more subjectivity into the equation. How exactly do we judge "progress"? Is the average person "happier" than they were 50 or 100 or 500 years ago? I'm not sure we have any great evidence of that. Murder rates and violent crime in general have mostly gone down over the centuries; is that "progress"? But as we make progress by some metrics, we falter by others. For example, child abductions have been declining for decades (especially by strangers), but we seem more terrified of stranger abductions than ever. Have we "solved" the social problem or merely traded the additional security of lowering already rare stranger abduction rates while increasing our general paranoid and fear level substantially (to the point that it disrupts child rearing and development)? Is that trade-off "progress"?
Historians have been talking about all of this for a lot longer than these tech books. Back in the old days, it was a debate over so-called Whiggish history [wikipedia.org]. Generally, that entailed an overuse of a teleological approach to historiography, or in simpler terms -- historians tended to write as though the present is the "goal" and the past is a set of gradual "progress" that "leads up" to that goal. The problem of course is that historical figures didn't know the future and they frequently had very different goals and priorities than we do today. It's questionable whether they'd even consider some of our narratives to be "progress" for the things they cared about. "Progress" is very relative to one's perspective, and these tech books are a similar critique to those historians have been making about narratives of "progress" for a much longer time.
One statement that stuck out to me in the review's conclusion:
I believe the implicit connection is laid bare in the implications of statements on the 'March For Science' website. The core principles state, again and again, that science "does things" for society. I think it's more accurate to say that people and technology (tools) do things, both positive and negative, for society--the fruits of scientific research merely provide means to make the tools.
Yes, that makes a lot of sense to me. Technology is about tools, and in some ways so is science (i.e., tools for understanding). But I still resist the author's need to try to superimpose arguments from these tech critics onto science. The reason the "March for Science" folks highlight technology is because "We have a better model for gravity" isn't going to mean much to most folks in their everyday lives. Technology allows a practical lens to see the possibilities created by progress in scientific knowledge. But whether that technology is "good" or "bad" or even constitutes "progress" in a social or cultural sense -- that seems to be confusing a bunch of categories.
(Score: 2) by pnkwarhall on Monday May 08 2017, @05:22PM (5 children)
First, thank you for your compliments. "Technopoly" is definitely worth another look--although it does focus on technological shifts, it contains an interesting perspective about how the mindset of "scientific inquiry as road to truth" led to the devaluation of that which could not be scientifically explored. I would also agree that the conflation of "science" and "technology" is wrong-headed--but my POV (as well as Postman's) is that this substitution/misunderstanding is not only the current general perspective, but a main source of societal problems.
Cultures and societies are the implemented solutions to fundamental human problems--Postman's main point is that a "technopoly" is a specific type of society (i.e. solution) to human problems that has serious negative consequences because it doesn't address, indeed ignores, vital parts of the human experience. Your statement that technology "allows a practical lens to see ["experience" is a more useful word, IMO] possibilities for progress is responded to directly by both authors with the assertion that, instead of "allowing" (i.e offering freedom) experience of possibilities, technological choices constrain the possibilities for human progress along particular paths. Lanier in particular offers concrete examples of more-or-less arbitrary software design choices that, due to widespread adoption of the specific software, became "baked into" not only further software development but actual cultural evolution and societal structure through the spread of fundamental memes with which we understand ourselves and our role in society.
I think it's interesting you mention that these ideas and criticisms are not new. (Postman's book starts with a reference to an ancient Egyptian argument about the societal value of the adoption of technologies.) One of my personal observations (particularly on Reddit) is that many people seem to believe that ancient cultures did not have similar debates, that in the modern era we are somehow vitally different (and advanced) in our perspectives because of our scientific and technological progress!
Lift Yr Skinny Fists Like Antennas to Heaven
(Score: 2) by AthanasiusKircher on Monday May 08 2017, @06:48PM (4 children)
Interesting. By the way, I absolutely agree with what you said about technology constraining as well as allowing possibilities of scientific application. My point in my previous post was restricted to the specific context about the "March for Science" that you mentioned in your review. I think the participants there were trying to emphasize practical application instead of pure scientific theory and knowledge, and that's what I meant by "allowing," i.e., the expansion from the theoretical to the practical. But you're right: the specific practical applications that end up becoming popular technologies then often constrain perspectives and possibilities.
I also briefly looked at some other reviews I could find on Technopoly, and I have to say that my memory of it was muddled and perhaps got a bit more confused by seeing your joint review along with the other book. As you note, Postman's critique is much broader than technology, and he is critiquing what is generally termed scientism [wikipedia.org], which both advocates for the expansion of "scientific" methods to all areas of human life and often argues against alternative (more "human" or "humanistic") approaches.
I sympathize with that view too. About 10-15 years ago I was involved in discussions about a couple projects that would now fall under what people call "digital humanities." At the time, I was kind of intrigued by the possibility of collecting "data" and manipulating it. I really thought the path forward to some traditional problems in that humanities field was basically through statistical analysis of large bodies of data.
Although my direct involvement with the project was brief, within a few years, such approaches started to become commonplace. And what result was, frankly, TERRIBLE. The very criticisms I had labeled at earlier research (arbitrariness, subjective judgments, focusing on the research author's interests or aims while excluding alternative possibilities) were simply translated into statistical methods. But now they appeared to have greater authority since they were based in "data" and had math to back them up.
I now realize that the earlier problems in the field weren't due to lack of "scientific" (or scientistic) rigor, but rather because research is always done by humans, and humans are flawed. Asking bad questions or using bad methodology with large datasets or statistics doesn't result in better results than asking the same bad questions without as much data or stats. Perhaps even more than technological concerns, to me the resonance with Postman (or at least from your review and the few others I just skimmed) is with the idea that information must be "the answer" to everything. But more information (omnipresent as a buzzword these days) just means it's even harder a create good interpretations of that information, which is often the key to understanding it.
So, yes, I agree with Postman's critiques of things like using IQ scores as a proxy for "intelligence" or assuming opinion polls really measure "what the public believes" rather than how the question was asked. Obviously a more quantitative and numerical approach to "human" problems can have benefits, but only if we recognize stats and data as one possible set of "tools" that can be used for good and bad too. It's interesting to think of Postman's criticism in relation to the recent controversies in many areas of science where results of major studies (that have been frequently cited) are then shown to actually not be reproducible when the experiments are run again. There are all sorts of flaws generally happening in these cases -- from confirmation bias in experimental design to data collection abnormalities to statistical errors to publication bias -- but at its heart what these things are pointing out again and again to me is that science is a human and social endeavor. And the human and social aspects are clearly not working well if we have lots of research being promoted that turns out to be wrong (because we don't encourage publication of negative results or reproducibility experiments or whatever).
Not only is science sometimes failing at being a good tool for understanding humanity, but the ejection of "human" concerns from science (or at least the failure to acknowledge their strong influence) has caused more holes to appear in actual scientific "progress" too.
(Score: 2) by pnkwarhall on Monday May 08 2017, @08:03PM (3 children)
I'm aware of "scientism", and thanks for pointing it out... I really should have included that term/concept in the piece.
Your point about "digital humanities" and statistics-based "science" is actually very close to one of the criticisms that led me down my current thought path. I was unaware until recently that there was this whole branch of scientific inquiry based around statistical inference, and my original criticisms of social science research were based on ideological principles. I guess there's a soylent commenter who consistently complains about 'null hypothesis statistical significance testing' (whose comments led me to look it up), and when I found out what it entailed, my principled skepticism seemed justified by the approach's reliance on way too many assumptions and simplifications.
Lift Yr Skinny Fists Like Antennas to Heaven
(Score: 2) by aristarchus on Tuesday May 09 2017, @08:37AM (2 children)
I was unaware until recently that there was this whole branch of scientific inquiry based around statistical inference, and my original criticisms of social science research were based on ideological principles.
Stats do not have ideology, unless they are fake stats, but then, those are not that hard to come by. .
I guess there's a soylent commenter who consistently complains about 'null hypothesis statistical significance testing' (whose comments led me to look it up), and when I found out what it entailed, my principled skepticism seemed justified by the approach's reliance on way too many assumptions and simplifications.
You should be equally skeptical of the "null hypothesis" skeptic, this seems to be a meme that is floating around these days, I hear it from people who listen to right-wing radio, and otherwise have no interest in science. Not saying the criticism is unfounded, but it is unfounded until it is understood.
But nice reviews, Pink Narwal!!
(Score: 1) by khallow on Tuesday May 09 2017, @02:07PM
Stats do not have ideology
Counterexample: finding the statistics that cast the issue in the best light for your ideology. For example, if you want to present the US's economy in a good light, speak of the US household average income of $52k per year. If you want to present it in a bad light, speak of the 38% of wealth owned by the 1%. Statistics even when valid is merely a viewpoint. And it is easy to choose advantageous viewpoints.
(Score: 1) by pnkwarhall on Wednesday May 10 2017, @11:04PM
>Pink Narwal
Thanks, I'll have to remember that one [i.redd.it].
Lift Yr Skinny Fists Like Antennas to Heaven