2 ms·

It's definitely a fuzzy distinction. Candidly, I'm skeptical about generalized methods that make claims about the future and pretend to do so rigorously. The i

It's definitely a fuzzy distinction. Candidly, I'm skeptical about generalized methods that make claims about the future and pretend to do so rigorously. The image that comes to mind when someone says "futurology" is those 19th century drawings of people riding flying bicycles to work: breathless excitement, comic in hindsight. To the extent that someone claims they can forecast something with so much uncertainty, the hair on my neck starts to stand up. And maybe I'm condemning a straw man, but it seems like this is your generalist-class, and most of the other groups who would not be as likely to self-identify as a futurist per se.

dredmorbius 5y ago
I found the quote I was looking for, it's in Arthur S. Iberal, Toward a General Science of Viable Systems (1972), and he's quoting Arnold Toynbee, A Study of History (1962). The passage concerns looking backwards rather than forwards, though the sentiment is similar: There is a strong tendency to depreciate works ... which are created by single mids, and the depreciation becomes the more emphatic the nearer such works approximate to being 'Universal Histories'. For example, Mr. H. G. Wells', The Outline of History was received with unmistakable hostility by a number of historical specialists. They criticized severely the errors which they discovered at the points where the writer, in his long journey through time and space, happened to traverse their tiny allotments. They seemed nnot to realize that, in re-living the entire life of Mankind as a single imaginative experience; Mr. Wells was achieving something which they themselves would hardly have dared to attempt -- something, perhaps, of which they had never conceived the possibility. In fact, the purpose and value of Mr. Wells' book seem to have been better appreciated by the general public than by the professional historians of the day. As to specificity of predictions, the example I like to use is of distinguishing between chronological predictions, that a specific even will happen on a specific schedule, and trigger or relational predictions, that one event or set of relationships predicates another. If a few minutes after 9 a.m. on the morning of 10 September 2001, you'd been standing on the corner of Liberty and West streets in lower Manhattan, looking north, and somebody were to ask you how much longer the two towers across the way would stand, your guess might have been decades to a century or more. If someone asked you the same question 24 hours later, your estimate might have been revised downwards considerably. Geology is a study which is remarkably imprecise in predicting when specific large-magnitude events will occur (earthquakes, volcanoes, landslides, etc.), though it has developed a very strong facility for suggesting where such indcidents are likely (and more pointedly, where not). The temporal uncertainty is a given due to the domain, characteristics, and timeline. The locational certainty is due largely to the predictability of what dynamics and structures give rise to events. (Yes, there are still surprises in both dimensions, though these tend to accrue additional knowledge.) All of the social and behavioural sciences have the handicap that the studies themselves exist within the system they are studying. Psychology, economics, sociology, political science, etc., operate within domains in which they are themselves among the variables and causal agents. Physical and biological sciences, for the most part, do not. This is seems to me a principle further confounding factor for the behavioural sciences, though it's not an insurmountable one. Even within those domains, activities and actions exist within constraint parameters, and eventually those constraints (one might call them ... limits) interpose. Peter Turchin is among those who's been doing interesting work in computational history, with some success.

dredmorbius 5y ago

I found the quote I was looking for, it's in Arthur S. Iberal, Toward a General Science of Viable Systems (1972), and he's quoting Arnold Toynbee, A Study of History (1962). The passage concerns looking backwards rather than forwards, though the sentiment is similar: There is a strong tendency to depreciate works ... which are created by single mids, and the depreciation becomes the more emphatic the nearer such works approximate to being 'Universal Histories'. For example, Mr. H. G. Wells', The Outline of History was received with unmistakable hostility by a number of historical specialists. They criticized severely the errors which they discovered at the points where the writer, in his long journey through time and space, happened to traverse their tiny allotments. They seemed nnot to realize that, in re-living the entire life of Mankind as a single imaginative experience; Mr. Wells was achieving something which they themselves would hardly have dared to attempt -- something, perhaps, of which they had never conceived the possibility. In fact, the purpose and value of Mr. Wells' book seem to have been better appreciated by the general public than by the professional historians of the day. As to specificity of predictions, the example I like to use is of distinguishing between chronological predictions, that a specific even will happen on a specific schedule, and trigger or relational predictions, that one event or set of relationships predicates another. If a few minutes after 9 a.m. on the morning of 10 September 2001, you'd been standing on the corner of Liberty and West streets in lower Manhattan, looking north, and somebody were to ask you how much longer the two towers across the way would stand, your guess might have been decades to a century or more. If someone asked you the same question 24 hours later, your estimate might have been revised downwards considerably. Geology is a study which is remarkably imprecise in predicting when specific large-magnitude events will occur (earthquakes, volcanoes, landslides, etc.), though it has developed a very strong facility for suggesting where such indcidents are likely (and more pointedly, where not). The temporal uncertainty is a given due to the domain, characteristics, and timeline. The locational certainty is due largely to the predictability of what dynamics and structures give rise to events. (Yes, there are still surprises in both dimensions, though these tend to accrue additional knowledge.) All of the social and behavioural sciences have the handicap that the studies themselves exist within the system they are studying. Psychology, economics, sociology, political science, etc., operate within domains in which they are themselves among the variables and causal agents. Physical and biological sciences, for the most part, do not. This is seems to me a principle further confounding factor for the behavioural sciences, though it's not an insurmountable one. Even within those domains, activities and actions exist within constraint parameters, and eventually those constraints (one might call them ... limits) interpose. Peter Turchin is among those who's been doing interesting work in computational history, with some success.