Realism and Simulation (was Open Letter...)
Matt DLM
I am responding to Cris Brown's "recapitulation" of our argument posted on
Monday 8 April. It is both amusing, because I see that much of the problem
has been simple miscommunication, and infuriating, because, once again,
everyone seems to be more interested in positing their reductionist
assumptions rather than my actual position.
You obviously have no understanding of my position. I'm not saying that in
a hostile way. The best way to understand where I coming from is to:
1. Read the introduction to Legacy of Glory. I'll be happy to send you
(or anyone else for that matter) a copy.
2. I wrote an article on LoG and design issues in EE&L (#8 I believe it
was).
3. For a different opinion, see Phil Johnston's excellent columns on this
very sort of debate in the last year of The Historical Gamer. He takes a
very reasonable approach in discussing both sides of the "realism" versus
"games" debate. In that, LoG was labeled a "perspective-based game", which
you will find corresponds rather amazingly with your definition of
"role-realism". (I'm glad to see that you have at last exposed yourself as
a realist after all!) Also, if anyone has copies of these articles, let
people know where they can get them.
There seem to be several ironies in your posting:
1. I don't think Jim Getz or Bob Jones have so-far been arguing for
"role-realism" or "perspective-based simulations" at all. They may join
in, but that hasn't been their tack so far.
2. By your new definition, you and I are in the same camp. Our only
disagreement would be the limits of "historical reasonbleness" or where a
warranted assertion falls on your "Q" scale. We even agree about the
benefit of having battalions represented if the player-perspective is of a
corps commander.
3. To say that someone is a "detail-realist" seems to basically say that
they can't see the forest for the trees. Clearly, anyone who tries to drop
this label on me makes it awfully easy to dismiss what I have been saying
and doing in this hobby. But I know it's a commonly repeated
misperception, but it's starting to feel like a slur because it is both
simplistic and false.
DETAIL is only important as it relates to perspective! You must have the
forest (role
-realism) before you can see the trees.
Cris, since we appear to be engaged in an increasingly productive
discussion, I hope you will take advantage of some of the publications I
mentioned. It will save everyone some time and frustration.
Matt
FX00
>>You've inverted the causality of my argument. I'll requote the last
sentence from my prior passage (above): "In my opinion, any grand-
tactical system which specifies a precise turn interval has problems,
BECAUSE THE HISTORICAL DATA IS SIMPLY TOO VARIED TO SQUEEZE INTO A
UNIFORM DECISION CYCLE."<<
>>Regardless of *why* commanders make decisions, the data indicate that
decision points tend to come in bunches, with long lulls between those
bunches. If you go with a long decision cycle (based on the lulls),
the simulation is historically unreasonable because it's impossible to
countermanuever in a timely manner. If you go with a short decision
cycle (based on the flurries), you have two options: (a) shackle the
players with strict C3I rules; or, (b) let the players change plans at
every decision cycle. The former makes for a boring game, because you
spend many turns simply shuffling units around, with no significant
decisions to be made. The latter gives rise to the "guided torpedo"
effect, and is thus an historically unreasonable simulation.<<
Can you give some pre-radio examples of these flurries. At the
grand-tactical level, I think the reaction time to enemy movements before
radio was at least half an hour, if not longer, due to transmission time.
For example, at Antietam, Jackson sent for Hood to come to his aid shortly
after Hooker's attack began (6:15-6:30), Hood begins his attack at about
7:00. In short, do you have examples of flurries that simply could not be
replicated in a set turn format.
>>You've begged the question. If grand-tactical commanders commonly
directed individual battalions, and if we aim to model the challenges
and opportunities which faced grand-tactical commanders (historically
reasonable) then the model should represent individual battalions. If
grand-tactical commanders almost never directed individual battalions,
then the model *shouldn't* represent individual battalions.<<
OK, if your SOLE purpose is to reflect the decisions of a single level of
commander (eg corps commander), then I concede your point. I think this
is not the case, however, in the vast majority of current games, which
generally simulate decisions made at several different levels of command
(for example, in Napoleons Battles, Army, Army Wing, Corps, Division and
Brigade).
>>Here I disagree wholeheartedly. The scientific method is founded upon
replicability. Experimental protocols are developed for the express
purpose of allowing other (and future) researchers to repeat the tests
and confirm the conclusions by first-hand observation. Where this is
not possible (e.g.: for transient observables), and where it cannot be
made possible by the recording of transient events for replicative
analysis, science turns its back and says "we don't know." (Or, more
often, science turns to skepticism and says "that didn't happen.")<<
>>History consists entirely of transient observables, and most of the
history with which gamers are concerned has not been recorded for
replicative analysis with anything approaching scientific rigor. The
vast majority of history is anecdotal, and more often than not the
reporters are selected (and then select what to report) according to
political and/or doctrinal criteria. And in most cases, that's the
limit of our access to the data; we can't run history again to see
whether Livy accurately chronicled the Punic Wars, or whether Ike was
lying when he said he and Bradley predicted the Ardennes Offensive.
All we can do is compare accounts and see who seems more reliable.<<
>>This is a *far* cry from the replicability inherent in the scientific
method. Science tends to (or at least claims to) withhold judgement
until there is proof beyond a reasonable doubt; history is satisfied
with a preponderance of the evidence, and often with far less. Thus,
to equate historical "facts" with scientific principles is a *major*
analytical error.<<
Hold on, Cris -- has the Big Bang ever been replicated? Has science
turned its back on this concept? I think not. Where has the evolution of
higher-order species been replicated? Any "we don't know" here? How
about sustainable nuclear fusion? Dark matter? I think you had better
rethink your position here.
The fact is, science can often replicate lower-order experiments, but many
broader concepts have NEVER been replicated. History can often replicate
lower-order experiments as well (for example, we can replicate the firing
of ACW artillery pieces).
>>No, history is the analysis of recorded *testimony*, complete with
biases, omissions, political and doctrinal dogma, puffery, myths,
rumors, misinformation, disinformation, and sometimes outright lies.
Assessing the reliability of that data is a daunting task, and we
ought never to assume that we've hit upon The Truth, merely because
we're able to find others (writers and readers) who agree with us.<<
Doesn't this also describe the often vicious battles that take place
around scientific controversies. The current war over the true nature of
the HIV virus springs to mind (a single virus vs. several related vires).
How about the inherent conflict between Newton and Einstein physics?
There are certainly differences between the evidence used by history and
science, but to argue that there is a catagorical difference does not bear
close scrutiny.
>>What I *have* argued is that there is no such thing as a *definitive,
conclusively correct* opinion regarding history. That is, there are
no opinions which are *so* well-researched and *so* well-reasoned that
no reasonable person could disagree.<<
What about my opinion that the Battle of Gettysburg took place in 1863 and
none of the combatants possessed nuclear weapon. Think I'll get any
reasonable person to disagree with me?
>>And, indeed, on many of the most
fundamental issues which confront a game designer, there are well-
researched and well-reasoned opinions to support opposing and mutually
exclusive conclusions.<<
Can you give examples of this? I believe it is an extremely rare
situation.
[THE SEQUEL]
>>Okay, FWIW I'm going to wax analytical for a few moments, and share
the six-point quantum of evidence scale I developed in my research on
experientialist epistemology. (BTW, this was fairly serious research;
I've presented and defended it in several doctoral-level seminars.)<<
[much cut]
>>So, where would we place most "historical facts" on this scale? I'd
suggest that the vast majority are no higher than a Q3 ("more likely
than not"). Worse, as we enter the level of detail necessary in a
tabletop model, the quantum of evidence slips still further. Few, if
any, of the "historical facts" most essential to a tabletop model are
greater than Q2 ("substantial possibilities"), and many are merely Q1
("reasonable possibilities"). I can think of *no* "historical fact"
which is significant from a designer's perspective and which I would
assess at Q5 ("beyond a reasonable doubt"), or even Q4 ("clear and
convincing evidence").<<
I will stipulate there are few Q5's. I think most are at Q3 or Q4,
depending on the game. But there are science Q2's, 3's and 4's as well.
Do you really think there is no clear and convincing evidence that, for
example, ACW infantry could move at least 1 mile in 30 minutes over open
terrain?
>>Note that I'm *not* saying the evidence is irrelevant. I'm saying the
evidence is *inconclusive*. (I've said this in several notes over the
past 48 hours, because it's become apparent that some folks think I'm
arguing for a nihilistic approach to history.)<<
Cris, I hate to break this to you, but to argue that all evidence in
inconclusive IS nihilism! If that's your position, I think you have to
live with the label.
Some have argued that it's unreasonable of me to demand a Q5, when
such evidence is unavailable. Fine. Don't claim your Q2 to be a Q5,
and I won't demand a Q5's worth of evidence. Admit that it's only a
Q2, and that's all I'll ask for as proof.
I'll argue that a Q3/Q4 is a Q3/Q4 and that reasonable conclusions can be
based on such evidence, because such evidence is accepted in the sciences
(and in every other field of human knowledge) all the time. Illustrative
example, but inapplicable in this case.
[ONCE MORE INTO THE BREACH, DEAR FRIENDS]
>>I'll give it a shot. It's a bit complicated, because there are three
issues involved. Two are at the surface, and provide the specific
context of the discussion. The third issue is one of policy, and is
the fire which fuels the debate. At the surface:<<
>>Matt (and others) are advocating what I call "detail-realism,"
where we measure the historical quality of a rules set by the degree
to which it gets the details right. Matt has suggested that we can
discuss which rules sets are "more realistic (historically accurate),"
where accuracy is "an accretion of details."<<
>>Jim and I (and others) are advocating what I call "role-realism,"
where we measure the historical quality of a rules set by whether the
overall flow of events gives the player the feeling that he/she has
stepped into an historical commander's role. I have suggested that we
can discuss which rules sets are "historically reasonable -- whether
the player is faced with challenges and opportunities appropriate to
the era and level of command, and whether the outcomes of decisions
are generally consistent with similar historical situations."<<
I'm not sure Matt has argued this issue from the point you ascribe to him.
I am certain I have not.
>>As a predicate issue, Jim and I (and others) have argued that it's
all but impossible to use Matt's detail-realism standard, because we
can't be that certain of the historical details. We argue that the
historical accounts are fraught with bias, omissions, misinformation,
disinformation, political and doctrinal dogma, and the like. I have
cited specific issues -- fundamental to game design -- upon which one
can make well-reasoned, well-researched arguments to support contra-
dictory and mutually exclusive conclusions, or where the historical
data are so varied that no uniform conclusion will be adequate.<<
>>Matt (and others) have suggested that Jim and I (and others) are
setting too high a standard for historical accuracy, and that we *can*
be certain enough of the details to make conclusive arguments as to
whether a rules set "accurately" represents those details.<<
I think this is essentially correct.
>>Thus, at the surface are debate over what might be termed an "inside-
out" (detail-realism) vs. "outside-in" (role-realism) approach to the
historical evaluation of tabletop models, and over whether we can have
sufficient historical certainty to use the detail-realism model. But
this is the *surface* debate; the fire something else altogether:<<
>>Matt (and others) have opined that "games" (which sacrifice histo-
rical accuracy in the interest of playability and fun) undermine the
hobby. Indeed, Matt has suggested "bad faith" on the part of those
who claim the term "historical" when they don't attempt to capture the
details of history.<<
>>Jim and I (and others) have rebelled against such an authoritative
attitude in a hobby. I've argued that -- since reasonable people may
reasonably disagree as to the history itself -- it is entirely likely
that two gamers may disagree as to which of the models is the better
presentation of history. Such disagreements do not give rise to the
inference that one of those gamers (the one with whom you disagree) is
ignorant of, or not interested in, history.<<
I have not argued this, either.
>>I suggest that the policy issue will not be resolved in this, or any
other, newsgroup debate. For this reflects an ongoing *societal*
debate which appears in our political, academic, and social discourse
on a regular basis: conservative criticisms of perspectivism, con-
structivism, experientialism, postmodernism, and other meta-analyses
which, in turn, criticise conservative authority. It is the never-
ending tension between order (we ought to acknowledge which games are
better historical models) and liberty (each gamer/designer has his/her
own ideas about history, and will choose/design games which reflect
his/her ideas).<<
An interesting analysis. I agree that the debate may (will) not be
resolved for either position, but I think it serves as a useful tool to
define the opposing positions.
Best
Fred Stovall
FX...@AOL.COM
Cris Brown
Cris here. :)
[Matt writes:]
>I am responding to Cris Brown's "recapitulation" of our argument posted on
>Monday 8 April. It is both amusing, because I see that much of the problem
>has been simple miscommunication, and infuriating, because, once again,
>everyone seems to be more interested in positing their reductionist
>assumptions rather than my actual position.
It may be that I'm misunderstanding you, but it may also be that
you're not saying precisely what you mean to say.
>You obviously have no understanding of my position. I'm not saying that in
>a hostile way. The best way to understand where I coming from is to:
>1. Read the introduction to Legacy of Glory. I'll be happy to send you
>(or anyone else for that matter) a copy.
I have it, and indeed I bought the game (over a year ago) because I
liked what you wrote in the intro. Then I read the rules.
Make no mistake about it; LoG is an *excellent* rules system. But I
think it has *far* more tactical detail than it needs, at the expense
of grand-tactical detail. I believe LoG is historically reasonable,
but I think ATOW-NB does a better job of reflecting what you wrote in
your introduction.
[...]
>3. For a different opinion, see Phil Johnston's excellent columns on this
>very sort of debate in the last year of The Historical Gamer. He takes a
>very reasonable approach in discussing both sides of the "realism" versus
>"games" debate. In that, LoG was labeled a "perspective-based game", which
>you will find corresponds rather amazingly with your definition of
>"role-realism". (I'm glad to see that you have at last exposed yourself as
>a realist after all!)[...]
I wouldn't read too much into that; I did use scare-quotes for "role-
realism," because I'm certainly *not* a "realist." (Without delving
any further into metaphysics, suffice it to say that I don't believe
there exists such a thing as "reality;" there are merely "collective
constructions of experience.") Unfortunately, "role-reasonability,"
"role-constructibility," and the like kinda flunk the music-of-the-
language test ... so I go with "role-realism."
[...]
>2. By your new definition, you and I are in the same camp. Our only
>disagreement would be the limits of "historical reasonbleness" or where
>a warranted assertion falls on your "Q" scale. We even agree about the
>benefit of having battalions represented if the player-perspective is of
>a corps commander.
I wouldn't be so sure; may major peeve with LoG is the presentation of
battalions. E.g., Rule 17.4.1: "The attacking player should spend no
time planning his move, units should simply be 'thrown forward' toward
the nearest enemy unit. (Players shouldn't get worked up about minor
tactical set-backs here or there -- that's not the concern of a
Player-Commander or the focus of a Grand-Tactical Operational game.)"
If battalion-level wrangling isn't the focus, and if I'm not supposed
to think about how I employ my battalions, *why include them at all*?
Again, I think LoG is an excellent rules set ... but this is a MAJOR
design flaw, IMO.
>3. To say that someone is a "detail-realist" seems to basically say
>that they can't see the forest for the trees. Clearly, anyone who tries
>to drop this label on me makes it awfully easy to dismiss what I have
>been saying and doing in this hobby. But I know it's a commonly repeated
>misperception, but it's starting to feel like a slur because it is both
>simplistic and false.
Oops. I *don't* equate "detail-realism" with not being able to see
the forest for the trees. Hmmm ... how to precisely explain what I
mean here....
Detail falls on a continuum; let's say there's a scale from 1 (very
abstract) to 10 (hyper-detailed). There seems to be a conceptual
membrane on the "detail continuum." If you stay on the low (abstract)
side of that membrane, you can lump a whole lot of stuff into key
abstractions and still preserve historical reasonability ("role-
realism"). Once you penetrate that membrane, the model demands more
details, and these in turn demand more details, in order to preserve
historical reasonability ("detail-realism"). Worse, as the level of
detail goes up, our confidence in the data usually diminishes.
I'll give a specific example. In BREAKTHROUGH (my WWII-era campaign
rules), I use an abstraction called Initiative Points (IPs). IPs
represent the availability and commitment of C3I and logistical re-
sources, operational flexibility, and marginal increments of time.
IPs spent on movement represent: scouting and planning routes of
march; keeping the roads clear of damaged vehicles and obstacles;
setting out traffic control checkpoints and refueling stations; and
the myriad of other administrative and logistical details which can
facilitate movement. Thus, you can move faster by spending more IPs,
but there's a tradeoff in that you'll expend your IPs faster as well,
which means your advance may stall before you reach your objective.
IPs spent on combat represent getting to the field a bit sooner, so
you can: reconnoiter enemy positions; pre-plot artillery and air
strikes; entrench; emplace or destroy obstacles and bridges; and get
your units supplied, organized, and disposed to fight. You can gain
significant tactical advantages by spending more IPs on pre-battle
preparation, but once again there's that tradeoff of operational pace.
IPs are one HUGE abstraction. But when I tried to break down the
components into separate factors, the details started to multiply
exponentially. I didn't feel I could design an historically reason-
able model of WWII-era operational art *without* considering the
administrative and logistical details, because my reading of history
suggests that these were often outcome determinative. But I couldn't
let the model *drown* in those details, because: (a) army and corps
commanders rarely attended *specifically* to those details; (b) it
makes for a boring game; and, (c) the greater the level of detail, the
lesser my confidence in the data.
As things stand, in BREAKTHROUGH you have to make choices of where to
commit your IPs. You'll usually have to maintain "quiet" sectors in
order to concentrate IPs elsewhere, or stand on the defensive for a
while in order to build up IPs for a major attack. You have to choose
between thorough pre-battle preparation and operational pace. You
also have to protect your "logistical tail" -- your IP Centers, your
transportation routes, and your Command and Formation HQs (where IPs
are "stored" until allocated, and spent, respectively). The result
is, I believe, an historically reasonable simulation of operational
art in WWII. But it's a very *abstract* simulation, because to get
any more concrete would have required an ocean of detail.
IMO, LoG penetrates that "membrane" in its tactical model, and is thus
awash in (unnecessary) tactical detail. That doesn't mean it's histo-
rically unreasonable, but it's certainly not to my taste.
Does this explain the distinction? If not, I'll try again.
Cris
"Every mighty oak was once a nut that stood its ground."
-- Anonymous
Cris Brown
Cris here. :)
[I argued that a uniform time cycle is unworkable, because my research
indicated that decision points tend to come in bunches, separated by
long lulls. Fred replies:]
>Can you give some pre-radio examples of these flurries. At the
>grand-tactical level, I think the reaction time to enemy movements
>before radio was at least half an hour, if not longer, due to trans-
>mission time. For example, at Antietam, Jackson sent for Hood to
>come to his aid shortly after Hooker's attack began (6:15-6:30),
>Hood begins his attack at about 7:00. In short, do you have exam-
>ples of flurries that simply could not be replicated in a set turn
>format.
Ahh ... I see that I've failed to describe what I mean, and thus we're
talking past each other. My apologies. I don't have my data at hand,
but I'll give you an idea of what it looked like:
0615 -- Issue orders to Units A, B, and D.
0630 -- Issue orders to Unit C.
0645 -- Issue orders to Units E and F.
LULL
0900 -- Issue orders to Units B and C.
0915 -- Issue orders to Units A, D, E, and F.
LULL
1230 -- Issue orders to Unit D.
1245 -- Issue orders to Units A, B, and E.
And so on. The lulls were times when the units were trying to carry
out the orders they'd just been issued: marching, fighting, etc. To
use a 15-minute turn cycle for this data would give lots of turns when
there were no significant decisions to make. To use two-hour turns
would lose much of the interplay. A one-hour turn is one of those
"worst of both worlds" compromises. Better to simply use variable-
length turns, based on what's happening on the field.
[Regarding unit scale:]
>OK, if your SOLE purpose is to reflect the decisions of a single level of
>commander (eg corps commander), then I concede your point. I think this
>is not the case, however, in the vast majority of current games, which
>generally simulate decisions made at several different levels of command
>(for example, in Napoleons Battles, Army, Army Wing, Corps, Division and
>Brigade).
I know, and that's one of my gripes with most of the current models.
You don't get a reasonable representation of command, because you are
simultaneously asked to be *too many commanders*. My preference is
for no more than two command perspectives, clearly delineated in the
mechanics, so that you always know which "hat" you're wearing.
[Regarding my emphasis on replicability as a distinction between
scientific and historical inquiry:]
>Hold on, Cris -- has the Big Bang ever been replicated? Has science
>turned its back on this concept? I think not. Where has the evolution of
>higher-order species been replicated? Any "we don't know" here? How
>about sustainable nuclear fusion? Dark matter? I think you had better
>rethink your position here.
Thanks for the advice, but I've thought about it quite a bit. ;)
Neither the Big Bang, nor higher-order evolution, nor dark matter is
treated as a "fact" in science. All of these are theories which seem
to fit the evidence we have available. As for sustainable nuclear
fusion ... we can all see the evidence of that every day, between
sunrise and sunset. The problem is that we haven't yet found a way to
replicate it in a way that won't blow us all up.
[I wrote:]
>>No, history is the analysis of recorded *testimony*, complete with
>>biases, omissions, political and doctrinal dogma, puffery, myths,
>>rumors, misinformation, disinformation, and sometimes outright lies.
>>Assessing the reliability of that data is a daunting task, and we
>>ought never to assume that we've hit upon The Truth, merely because
>>we're able to find others (writers and readers) who agree with us.
[Fred replies:]
>Doesn't this also describe the often vicious battles that take place
>around scientific controversies. The current war over the true nature of
>the HIV virus springs to mind (a single virus vs. several related vires).
Note, again, that these "scientific controversies" surround theories,
and not things which have been labeled "laws," or "facts." The big
difference is that lay historians (i.e.: the general reading public)
tend to treat history as inarguable *fact*, rather than *theory*.
>How about the inherent conflict between Newton and Einstein physics?
What conflict? Newtonian mechanics and Einsteinian relativity apply
to very different domains, and thus there's no reason to think they
ought to behave the same way. You're buying into the common fallacy
of reductionism (see Leshan & Margeneau, EINSTEIN'S SPACE AND VAN
GOGH'S SKY (1982) at 92-129).
>>What I *have* argued is that there is no such thing as a *definitive,
>>conclusively correct* opinion regarding history. That is, there are
>>no opinions which are *so* well-researched and *so* well-reasoned that
>>no reasonable person could disagree.
>What about my opinion that the Battle of Gettysburg took place in 1863
>and none of the combatants possessed nuclear weapon. Think I'll get any
>reasonable person to disagree with me?
Here the fallacy is an error of scope. This discussion is about the
kinds of historical conclusions which are needed to design a tabletop
model; your proposed counterexample is irrelevant, as it lies outside
the scope of the discussion.
>>And, indeed, on many of the most fundamental issues which confront
>>a game designer, there are well-researched and well-reasoned opinions
>>to support opposing and mutually exclusive conclusions.
>Can you give examples of this? I believe it is an extremely rare
>situation.
Sure. How effective was the bayonet in the Napoleonic era? Was there
a significant tactical distinction between a battalion in column and a
battalion in line? Had cavalry been rendered battlefield-ineffective
by the ACW, or did ACW commanders simply fail to employ cavalry to its
best effect? What were the relative combat effectiveness of infantry,
armor, artillery, and airpower in WWII? Had airpower become the most
decisive combat arm by DESERT STORM?
*All* of these are issues upon which reasonable people may reasonably
disagree, grounding their arguments in the historical data.
[Regarding my Q-scale for evaluating evidence:]
>I will stipulate there are few Q5's. I think most are at Q3 or Q4,
>depending on the game. But there are science Q2's, 3's and 4's as well.
>Do you really think there is no clear and convincing evidence that, for
>example, ACW infantry could move at least 1 mile in 30 minutes over open
>terrain?
Show me a model which postulates infantry movement in the ACW as "at
least one mile per 30 minutes, over open terrain" and you've got an
argument. Most of the models I've seen are a tad more precise. And
it's that need for precision which drops the Q-value.
>>Note that I'm *not* saying the evidence is irrelevant. I'm saying the
>>evidence is *inconclusive*. (I've said this in several notes over the
>>past 48 hours, because it's become apparent that some folks think I'm
>>arguing for a nihilistic approach to history.)
>Cris, I hate to break this to you, but to argue that all evidence in
>inconclusive IS nihilism! If that's your position, I think you have to
>live with the label.
This is a fallacy known as "argument by false extremes." Authori-
tarianism and nihilism are not a binary set. They lie on a continuum,
with intermediate values. My failure to adopt your extreme does not
place me at the opposite extreme.
>>Some have argued that it's unreasonable of me to demand a Q5, when
>>such evidence is unavailable. Fine. Don't claim your Q2 to be a Q5,
>>and I won't demand a Q5's worth of evidence. Admit that it's only a
>>Q2, and that's all I'll ask for as proof.
>I'll argue that a Q3/Q4 is a Q3/Q4 and that reasonable conclusions can be
>based on such evidence, because such evidence is accepted in the sciences
>(and in every other field of human knowledge) all the time. Illustrative
>example, but inapplicable in this case.
If you can defend a Q3 as a Q3, that's fine. But don't claim it to be
a Q5, permitting no reasonable alternative. The problem isn't that we
don't know anything. The problem is that we almost never know so much
that we can tell someone else "You don't know anything."
Engcyb0
Can we all agree on one thread to post in for this discussion? Please
please?
I am getting confused trying to follow it through so many other threads.
May I propose starting a new thread and posting everything from the four
threads in question there?
CB
engcyb0