[lit-ideas] Re: Hitchens Arguably on John Brown

  • From: Donal McEvoy <donalmcevoyuk@xxxxxxxxxxx>
  • To: "lit-ideas@xxxxxxxxxxxxx" <lit-ideas@xxxxxxxxxxxxx>
  • Date: Tue, 18 Oct 2011 11:50:04 +0100 (BST)

The probability of any event E (including an event that is a combination of two 
or more events, or what we might sometimes term a 'coincidence'] will lie 
somewhere between 0 (which denotes the event has zero probability and therefore 
is impossible) and 1 (which denotes maximal probability and therefore that E is 

The probability of any E over an almost infinity of time is vastly increased 
compared to its probability over a relatively short period of time (unless we 
bring in further assumptions, including ruling out E as impossible at any 
time), and E may approach being almost certain (unless it is ruled out as 
impossible) as any possible event may be thought to be almost certain to occur 
given enough time, and infinite time is surely enough. We use this principle in 
calculating the probability, say, of someone breaking their leg in the course 
of their life as opposed to before the age of ten, knowing that this 
probability generally increases over time because there are more opportunities 
(as it were) for a leg-breaking E to occur. [It was therefore very significant 
for researchers to find that the occurrence of depression in persons born after 
the war was much greater than that in persons born at the beginning of the 
century: this showed depression was both
 occurring earlier in individuals and striking more often with those born 
later, a marked 'change in probabilities' that therefore invites explanation].

This is enough to indicate that the 'probability of E' may vary in accordance 
with other factors, like the parameters of time involved and the state of 
affairs (or other events) against which that probability is measured: the 
probability of a mass marketed home computer existing by 1990 would have varied 
if the question were asked as if we were back in the year 1066, the year 1666, 
the year 1966 and the year 1976: only in a certain form of determinism would we 
say that it was always preordained that there would a mass marketed home 
computer by 1990 and therefore this E was certain [with probability 1] at all 
stages in history, and that it is simply because probability measures our 
subjective lack of complete knowledge that this probability varies with changes 
in our subjective lack of knowledge over the centuries.

When we speak of a coincidence, even a 'pure coincidence' or a 'staggering 
coincidence', we are all either talking loosely or we are straying into an area 
that requires understanding of probabilities and indeed the various 
'philosophies' of probability. It is a further question how the logical 
analysis of the 'probability of E' ties in with our psychological sense of the 
liklihood or otherwise of E; but we need the logical analysis to put our 
psychological intuitions into rational perspective.

For example, take an intuitively plausible claim like,"Coincidences ONLY exist 
because of the miniscule chance of some things 
overlapping in the trillions of things that happen in a given time 
period." For reasons indicated "a given time period" may affect probabilities 
(and therefore "coincidences" where these are seen as improbable occurrences), 
but it is nevertheless way too simple to say coincidences or improbable events 
"ONLY exist because of the miniscule chance of some things 
overlapping in the trillions of things that happen in a given time 
period". It may be true that, if we are not determinists, then it is probable 
over "a given time period" that some improbable events, even highly improbable 
events, will occur: but this does not explain why these occur while many more 
"improbable events" do not; or why these improbable events occur and many 
probable events do not in fact occur [for every highly improbable event that 
occurs, we might say, there must be at least one more-probable event that has 
therefore not occurred]. We may say, almost by definition or a priori, that it 
is probable that a greater percentage of highly probable events will occur 
[relative to those that do not] than highly improbable events will occur 
[relative to those that do not]. But this, even if true, does not explain 
itself. Still less does it help explain why some highly improbable events occur 
and some do not. Nor does it allow us to conclude they "ONLY exist because" of 

In fact, it is unclear what theory or philosophy of probability the claim, 
"Coincidences ONLY exist because of the miniscule chance of some things 
overlapping in the trillions of things that happen in a given time 
period", is meant to reflect or assert. This is a weakness in such a claim. If 
such a claim is meant only to express a deterministic POV, such that 
events/coincidence only appear improbable but in fact are not, then this may be 
discussed on its own terms.

[At his death, in his nineties, probability theory was one of the topics Popper 
was still working on btw, and he was defending an objectivist account of 
probabilities as measures of underlying propensities in states-of-affairs for 
changed states-of-affairs, as part of understanding the universe as involving 
"changed propensities for change". Probability theory remains an intellectual 
minefield even if we progress beyond our untutored intuitions.]

Where snow is falling like pieces of a snowman that had just stepped on a mine 
in a minefield

From: Eric Yost <mr.eric.yost@xxxxxxxxx>
To: lit-ideas@xxxxxxxxxxxxx
Sent: Tuesday, 18 October 2011, 7:17
Subject: [lit-ideas] Re: Hitchens Arguably on John Brown

Colin Bruce’s _Conned again, Watson!_ describes many of the perceptual biases 
that people read into, hope for, or are manipulated by in daily probabilities. 
He agrees with Paul.
From:lit-ideas-bounce@xxxxxxxxxxxxx [mailto:lit-ideas-bounce@xxxxxxxxxxxxx] On 
Behalf Of Paul Stone
Sent: Monday, October 17, 2011 9:06 PM
To: lit-ideas@xxxxxxxxxxxxx
Subject: [lit-ideas] Re: Hitchens Arguably on John Brown
No magic!


Other related posts: