[EM] James, 5 April, '05, 0645 GMT

[MIKE OSSIPOFF]

James--

You said:

This is James G-A replying to Mike Ossipoff, on the topic of strategic
voting... (When I write "James:" before a bunch of quoted text, that means
that the text was originally written by me, and quoted by Mike. When I
write "Mike:" before a bunch of quoted text, that means that I am quoting
Mike.)

As I said, I'll adopt that usage, for these messages, except that I'll also
write "Mike:" in front of my current reply, so that won't be confused with 
what
I'm replying to.

Mike:

>To bury something means to put it under something else, or to conceal it.
That's
>exactly the meaning of burying that I use when I refer to favorite-burial
>strategy in Plurality or IRV.
>If that isn't burying to Blake or to you, then you're using a different 
>meaning
>of burying, different from what the word means to others.
>But I have no complaint if you or Blake invent new meanings for words.
>

James:

	Burying means to place one thing under something else.


Mike:

Yes, James, that's what I just finished telling you.

James:

In
this case, I argue that it makes less sense to say that you are burying
Nader and Camejo under Kerry, and more sense to say that you are raising
Kerry above Nader and Camejo.

Mike:

You're doing both, James.

To say either would be completely correct. For some reason, however, you 
have a
problem with the fact that I don't say what you say.

James:

This is a useful distinction that Blake's
terminology helps us make.
	Why does it make more sense to say this? Because if you are "burying"
Nader, it seems to imply that your purpose is to change the probability of
Nader's being elected.

Mike:

Then it seems to you wrong. Millions of voters bury Nader in order to help
Kerry. Are they also raising Kerry? Yes. Do you have a right to call it what 
you
want? Yes, but you're wrong when you say that you're somehow more accurate.

The definition of "bury" that you quoted says nothing about intention. Feel 
free
to make up what you want, but try to distinguish between added meaning that 
you
make up, and meaning that was already there.

James:

Actually, that isn't your purpose, although it may
be a side effect. Your purpose is to improve Kerry's chance of being
elected (thus decreasing Bush's chance of being elected), hence it makes
sense to say that you are raising Kerry, rather than burying Nader.

Mike:

Don't insult me by implying that I'd vote Kerry over Nader. If you do, I'll 
call
you James A.

So let's talk about why _you_ voted Kerry over Nader last November. You did 
it
to help Kerry rather than to hurt Nader, and so, to you, what you're doing 
to
Kerry is more important than what you're doing to Nader. So, understandably,
you'd rather call that raising Kerry than burying Nader. Fine. But you're 
still
doing both.

I choose to talk about and emphasize the  burial of favorites, because I
consider that to be an important problem.

It's ok for you to say that you're raising Kerry and not mention the fact 
that
you're burying Nader, but that doesn't mean that you aren't burying him by 
the
ordinary general meaning of "bury". But you must try to understand that it's
also ok to speak of the burying of your favorite, if somone other than you
considers that to be an important problem.

Mike:
>Thank you for that grounding in the basics. :-) If Blake's website is the
>basics, then I fully admit that I'm not grounded in the basics.
>

James:

	No, Blake's website in itself isn't "the basics". But it contains many
useful definitions

Mike:

Surely most any definition is useful for something. Whether someone's
definitions for the things that he discusses are useful for discussion of
different things is another matter. Why don't I use Blake's definitions? 
Because
the things that he defines and discusses aren't of interest to me.

James:

Mike:
>
>I've repeated this very many times, including a few times during the last 
>few
days, but apparently it's necessary to repeat it again for you:
>Your emphasis on methods' immunity to strategy is barking up the wrong
>tree. Plruality is immune to what you call burying strategy. So what? The
>reason why Plurality is widely recognized as the worst is because, though 
>it is
immune to what you call burying strategy, Plurality, given the situation 
that
millions of voters believe exists, strategically forces voters to
>bury their favorite.
>The problem is voters' need to reverse preferences or abandon or bury
>their favorite. That's why Plurality is widely recognized to have an 
>especially
serious strategy problem even though Plurality is immune to what you call 
buring
strategy.
>I'm not interested in immunity to what you call burying strategy. I'm not
interested in methods susceptibility or immunity to strategy. I'm
>interested in the extent to which they strategically force voters to 
>abandon
their favorite and conceal what they really want.
>I find that it's necessary to repeat the same things so often that I
>should just write these answers into a FAQ, and post the FAQ every time you 
>or
someone like you repeats the same questions and mis-statements.

James:

	That's not necessary in this case. I was already aware of your point of
view regarding the relative importance of "offensive" versus "defensive"
strategy. (I also get annoyed when it seems I am repeating myself more
often than necessary.) However, the fact that I understand your point of
view doesn't necessarily mean that I agree with it.

Mike:

If I'd believed that you agreed with me, would I have said that you're 
barking
up the wrong tree?

James:

Translating your argument into terms that I can readily work with, you
seem to be saying that we shouldn't worry about the burying strategy in
itself

Mike:

Correct.

James:


, but only about the compromising strategy

Mike:

No. For one thing, the problem isn't the defensive strategy, but rather the
problem is the _need_ for defensive strategy.

Well ok, the defensive strategy itself is bad news of course, when it 
conceals
what people want, but it seems more productive to speak of the goal of 
getting
rid of the need for it.

For another thing, I've made it clear again and again that "compromising" 
out of
defensive strategic need is the problem, not compromising per se. It's true 
that
I proposed AERLO initially for the purpose of getting rid of a mere 
incentive,
as opposed to defensive need, for equal ranking. But the majority defensive
strategy criteria are about the goal of minimizing defensive need.

Dealing with defensive need, the worst strategy problem, is my more modest 
goal.

James:

..., and the
counterstrategies needed to fight the burying strategy.

Mike:

No, not unless those counterstrategies are for protecting the win of a CW or
protecting majority wishes.

My majority defensive strategy criteria describe condidtions under which a
majority won't need certain strategies.

Use whatever terminology you want to, but understand that as you quote what 
I've
said back to me in Blake's terminology, you've got the meanings all changed,
because Blakes terms name what Blake discusses, not what I discuss.

James:

	First of all, I agree with you that the compromising strategy is a
serious problem.

Mike:

It isn't, unless reversing a preference or burying a favorite is needed in 
order
to protect the win of a CW or to protect majority wishes.

James:

Plurality has an extremely nasty compromising strategy
incentive

Mike:

If Plurality only had a strategy incentive, as opposed to a defensive 
strategy
need, and if it were only an incentive for
what you call compromising, rather than for defensive favorite-burial, then
Plurality wouldn't have a serious or nasty problem at all.

James:

In short, I strongly agree with your goal to minimize the
compromising strategy.

Mike:

Well, I suppose you could say that if my goal is minimizing the need for
defensive order-reversal, then that implies a goal of minimizing the 
defensive
order-reversal itself. But that's getting a little ahead of ourselves. Let's
just say that my goal is to minimize the need for defensive strategy.But 
when
you say that my goal is to minimize all "compromising", you're naming a goal
that's a little more broad than the goal of my majority defensive strategy
criteria.

I'll say it again: Blakes terms just don't describe the things that I 
discuss.

James:

	We can observe the harmful effects of the compromising strategy, and we
can see the need to minimize the incentive for it. But what about the
burying strategy? Where can we observe its harmful effects? The answer is
that we can't, because methods that are susceptible to it are not
currently used for major public elections. Hence, the burying strategy is
not a problem in elections today, but it might become a problem if
Condorcet methods (or approval, Bucklin, etc.)

Mike:

Could you post an example of how your "burying" strategy would work in 
Bucklin
and Approval?

James:

...are adopted. In order to
understand the harmful effects of the burying strategy before
burying-vulnerable methods are adopted, we need to use our foresight, our
imagination. We need to take steps to prevent serious burying problems
before they occur, or risk discrediting Condorcet systems in general.
	Why, in your opinion, do we not need to worry about the burying strategy?
If a group of voters is able to change the outcome to the ideological
polar opposite of the sincere CW by means of a burying strategy, is this
not a serious problem? Of course it is.

Mike:

It is if you insist on concentrating on it as a separate strategy problem. 
Look,
whenever offensive order-reversal could be used, it could be protected 
against
by some defensive strategy.

In Margins, the needed defensive strategy could be defensive 
favorite-burial. In
Condorcet, defensive equal ranking can thwart offensive order-reversal, or
deterrent truncation could discourage the use of the offensive 
order-reversal.

But do you really think that it's an important distinction whether that need 
for
defensive order-reversal is caused by Margins' offensive strategy, or 
whether
that defensive order-reversal need exists in Plurality or IRV without any 
need
for offensive strategy? Same problem, James.

So you can forget about insisting that what you call "burying strategy" 
makes a
distinct problem, a problem distinct from the problems of methods that don't
have what you call a "burying" strategy.

Your "burying strategy" is just one mechanism of  more general problem that
exists in methods that don't have what you call "burying strategy".

James:

Can we take steps to prevent it?
Yes, we probably can in most cases, but in doing so, we acknowledge that
the burying strategy is worth worrying about.

Mike:

Sure, but not for itself. As a mechanism for a more general problem, but not 
as
a problem in its own right.

James:

This brings us to the issue
of counterstrategy. What counterstrategies can voters use to try to make
burying strategies unprofitable? How effective are these strategies, and
how risky are they in their own right?
	These are questions that we have discussed in detail already, so I know
that you are at least somewhat interested in them. If you are able to show
that particular Condorcet methods are not particularly vulnerable to the
burying strategy, that's great, but it doesn't mean that the question of
vulnerability itself is unimportant ("barking up the wrong tree").

Mike:

It's barking up the wrong tree, because you're speaking of one mechanism of 
a
more general problem as if it were a separate problem in its own right. The
whole notion of "vulnerability to strategy" is barking up the wrong tree, 
and
amounts to completely missing the point.

When someone speaks of Plurality's vulnerability to compromise strategy, 
that
undescores and dramatizes the silliness of that approach, even if you 
yourself
don't word it quite that way.

James:

	Condorcet methods (in place of plurality, runoff, IRV) minimize the
compromising strategy, but introduce the burying strategy.

Mike:

No, Condorcet doesn't add any problem. Condorcet wv only has defensive 
strategy
need when offensive order-reversal is used or expected. But even then, 
Condorcet
wv doesn't give a need for defensive order-reversal. Even in its worst-case,
Condorcet doesn't have anything like the defensive strategy need of 
Plurality,
Runoff, and IRV.

You'd gain some clarity if you'd capitalize the names of voting systems and
criteria. Style doesn't permit it? Then you'd better obey style.

James:

As long as we
choose Condorcet methods that are relatively good at keeping burying under
control (e.g. not margins!), I argue that this tradeoff is clearly
worthwhile. However, when arguing for the adoption of Condorcet, we must
recognize that this is at issue. If we conceal the issue, I argue that we
are just as intellectually dishonest as IRV advocates who advertise that
IRV "eliminates the spoiler problem" and "ensures majority rule".


Mike:

Condorcet wv doesn't add a problem. It has a vestige of the problems of 
Plurity,
IRV, and Margins, greatly reduced, effectively eliminated.

>
>
James:
>

>	In that post (one year ago tomorrow) I discussed the difference between
>your criteria and Blake's criteria in some depth.


Mike:

Good. No doubt you discussed this difference: My criteria are
universally-applicable, and Blake's Condorcet Criterion, and some of his 
others,
explictly limit their applicability to rank methods. Not very useful? That's
right.

Mike:
>
>You know, to discuss the difference between my criteria and Blakes, I'd
>have to be conversant abouit Blake's criteria. I can't say that I am.

James--

	I can see no good reason for you not to be. If you are interested in
engaging in discussion on this list, that sometimes means talking about
someone else's ideas besides your own.

Mike:

You're talking like an ass. You're suggesting that I never talk about 
anyone's ideas other than my own. I've talked about your ideas, for 
instance. I've often replied to various ideas expressed by various 
list-members. Where have you been?

I said that I'm not conversant about Blake's criteria. That's because they 
don't interest me. But it doesn't mean that I haven't discussed them at some 
time in the past. Long before the time when you say you began to study 
voting systems, I'd copied and pasted Blake's criteria into a posting to EM, 
and discussed his criteria at great length.

I devoted a posting to Woodall's criteria. ("Woodall's Whacky & Zany 
Criteria").

Mike Ossipoff

_________________________________________________________________
Express yourself instantly with MSN Messenger! Download today - it's FREE! 
http://messenger.msn.click-url.com/go/onm00200471ave/direct/01/