h2g2 The Hitchhiker's Guide to the Galaxy: Earth Edition

I'm just coming to the end of the excellent 'Guns, Germs and Steel' by Jared Diamond. In the last chapter he's justifying the study of History as a science and says some very interesting things about 'science'.

Physics is sometimes taken as the canonical model of 'proper' science. In physics you can set up carefully controlled experiments and make hard and fast predictions. The physicists tend to be sniffy about the less precise sciences.

However, the physics like model works in very few domains (physics, chemistry, molecular biology). In what Diamond calls the 'historical' sciences - he includes astronomy amongst them - we have to work in complex, multivariate, uncontrolled environments. Starting conditions are always unknowable (since we don't know everything that has gone before), we can only make messy, approximate inferences and we can seldom set up neatly controlled experiments. Thus alternative methodologies are needed. Explanation rather than prediction. Statistical analyses. Comparative studies.

So, yes...slightly different ways of thinking are needed for different types of problem. But they're all fundamentally grounded in empiricism, no?

Even in 'historical' sciences, it's still possible to make predictions about what will be found when you look somewhere you haven't looked before (deeper into the universe, using a new wavelength, etc).

True. It's a bit of a fuzzy distinction, though (which is basically my point: they're all the same kind of science). If you know that a star has a particular visual spectrum, *physics* allows you to predict its radio spectrum. However...pure physics-based extrapolation will not tell you how far away the next star is or what its spectrum will be.

To use another example closer to Jared Diamond's subject matter:
We can explain why technological civilisations (and their accompanying infectious diseases) came into being in Eurasia sooner than in the Andes and thus allowed the Spaniards to conquer the Incas, rather than vice versa. What we *can't* tell is what Pissarro had for breakfast, why he didn't succumb to yellow fever, how things might have gone if he had...etc.

And another example, having just completed the last couple of pages:
We can say with reasonable certainty that in a given maternity hospital, of the next 1000 babies born, between 480 and 520 will be boys. But we can't tell what the next 2 babies will be.

(For avoidance of pedantry: This would only be true in the absence of 'treatment variables' - eg it might not apply in a region where selective abortion of girls is practiced).

That last thing is nothing to do with the difference between sciences like physics and subjects like history. Take thermodynamics or quantum physics for example. It's just probability and statistics.

Absolutely! That's my point. It's not 'another way of thinking.'

'Probibalistic logic', indeed. Pshaw!

I rather think it was you who introduced the phrase "another way of thinking". I have (I hope!) always recommended 'horses for courses' - the appropriate tool for the job. More specifically, I have spoken out against using the wrong tool for the job, and maybe pretending it was OK to do so, even knowingly.

In a field such as quantum mechanics, for example, wouldn't you say that probabilistic logic would be employed? A TRUE/FALSE logic would be inappropriate and unhelpful, I would've thought. Pshaw yourself!

Pattern-chaser

"Who cares, wins"

No. The mathematics on which probability/ statistics is based is founded on the same principals of logic. One or Zero. True or False. Much else follows.

Russell et al made as stab at showing how *all* mathematics can be traced back to basic logic in the Principia Mathematica. They abandoned the project with the advent of Goedel's Theorem - that some things are unprovable - and we can't tell in advance whether or not they are...but that's another story.

(An illustration of Goedel's Theorem that I thought up the other day:
We have no way of knowing what Napoleon ate for breakfast on any given day. However...his diaries might turn up.)

Are these principals like Atlas? Do they get regular breaks? ...I couldn't resist it.

Seriously, is this just semantics again? Doesn't statistics set the rules for how we may manipulate and combine probabilities, so as to arrive at valid, meaningful and useful conclusions? Isn't that more or less the definition of a probabilistic logic?

Pattern-chaser

"Who cares, wins"

If you like.

But remember again why we're talking about the distinction. Bodywold or Mindworld. Way back when you seemed to be saying (something like):
- If we look from the viewpoint of the physical world, we can apply Boolean logic.
- If we look from the perspective of our internal world, we need probabilistic logic.

And I kept saying that the Mindworld was part of the Bodyworld. And now I'm saying that Probibalistic logic (if you insist) is the same kind of beast as Boolean logic.

Hence my 'Pshaw!'

Anyway...it wasn't me going 'Pshaw!'. You're imagining me.

<- If we look from the viewpoint of the physical world, we can apply Boolean logic.
- If we look from the perspective of our internal world, we need probabilistic logic.>

I think it was the other way around, but who cares? I've been thinking about this. If my description of perception is correct, then the only valid () reasoning we can do concerning the Bw is speculation. On the one hand, this seems a bit pointless. But on the other, the alternative is that we can't consider the Bw directly at all, but only indirectly via our model. And that would mean that learning something - anything! - new would be difficult if not impossible. We could refine what we know, but we couldn't make any significant additions.

What sent me off down this line of thought is the way our perception fills in any 'gaps' for us. What does it fill them with? With whatever our model says should be there. I.e. what we perceive is our world model; what we expect to see. [Although we believe that the real world underlies it, and resembles it to some unknown but hopefully significant degree.]

We actually live in the Mw, or that portion of it that contains our personal world map. Which brings me back to "equal rights for Holmes", but let's consider the above first. I take it you - dear Reader - probably don't agree, which is fine. Can you identify any areas where you think I'm mistaken?

Pattern-chaser

"Who cares, wins"

'Speculation' isn't a different a *form* of reasoning.
Depending on context, it's merely an acceptance that the basis and/or results of reasoning are uncertain, or an arbitrary qualitative category used for *relatively* uncertain reasoning or results.

*Describing* reasoning about reality as speculative seems relatively pointless if all *reasoning* is going to be described that way.
Once people accept they can't really be certain about anything, it seems more use to have terms which are useful for describing the *relative* reliability of different conclusions than merely repeatedly restating what people already know just in case they might forget.
Words that cover all eventualities have limited, if any, descriptive value.

The point in people having a practical scale of description from, say, 'pure speculation' to 'certainty' is to be able to describe a whole range of reasoning about the real world, not to be taken mindlessly literally.
Few people are daft enough to believe that if someone claims their conclusions are perfectly correct, they actually *are* perfectly correct - most would understand that either the other person *thinks* they are correct, or wants other people to think they are.

>>"What sent me off down this line of thought is the way our perception fills in any 'gaps' for us."

We constantly 'fill in' for things we can't observe because they're out of the range of our perceptions by the simple expedient of extrapolating the past into the future.
We assume that much of the material world simply stays where it is, that moving objects move according to one of a few simple rules (constant velocity, parabolic flight, slowing under friction, etc), that certain objects age at various rates or, more generally, that chemical reactions have some 'momentum' and 'direction'.

When our 'predictions' turn out wrong, the more curious among us wonder why, and try to develop better mechanisms for 'filling in' in space and time.
Arguably the goal of much learning is precisely to be able to fill-in accurately - if the evolutionary point of a mind is to be 'a machine for producing future', *temporal* 'filling in' is the basic point of the brain, and a huge evolutionary driver.

Unless we're pretty ****ed-up, we don't live purely in the mind-world, or at best, we live with a very thin skin seperating us from much of reality. If we mistake the position of a door in a wall, or the floor under our feet, most of us pretty soon realise our mistake. It's only possible to be sane and live in an *undisturbed* Mw for long if our world model is already pretty accurate within the range of our likely physical actions

>>"Can you identify any areas where you think I'm mistaken?"

Can you identify any areas where your approach gives some new perspective?

This place has died since P-c left hasn't it?

Anyway, just started a forum conversation that might be of some interest.

F135418?thread=3473890?thread=&skip=0&show=20

Until now, I hadn't realised he'd left 'as such', though I have been distracted the last month or so.
I kind of thought the conversation had pretty much come to a natural end.

Looking around, it seems there was stuff happening I wasn't aware of, which I don't understand, and which it's probably better I don't try and understand.

I missed the drama too, Potholer.

I suppose when a conversation becomes reduced to irrelevant minutiae and tired useless concepts, and then source of said irritants is removed, the conversation is at an end...

Ste

That's unfair Ste.

Potholer, yeah it looks like he left over something apparently quite minor. Perhaps he'll be back when he reflects on it.

Name:The_Doctor_Is_In

Chair title:The Tex Avery Directorate of experimentation on non conformist hip swinging pensioners with mindless disregard for authority figures wearing Pizza hats

Any beliefs you'd like to list so we can make fun- er... discuss them:

My beliefs are that there is no such thing as evil breasts, that it is every mans right to dress up as a French Maid at least once in his life and that every company in the land declare Fridays as "Gerry Anderson" day, whereapon all workers will dress as their favourite GA character and walk about like Supermarrionated puppets.
There should also be enforced watching of the "Green Wing" TV show for all non believers.......