Utilitarianism and Justice

[Eric Sanders]

Which would you prefer:

A great education for 1/2 the students and a poor education for the other 1/2?

OR

An okay education for all the students?

Is there a formula that could help us decide which is more just?

[Dale Sheldon-Hess]

What are the utilities of a great, okay, and poor education?

If the utility of an okay education is closer to great than to poor,
than okay for all has higher total utility.

--
Dale

[Warren D Smith]

>> Which would you prefer:
>>
>> A great education for 1/2 the students and a poor education for the other
>> 1/2?
>> OR
>> An okay education for all the students?
>>
>> Is there a formula that could help us decide which is more just?

--interesting.
One could argue that (maybe) society is going to be way worse if NOBODY gets a
really great education. (No scientific progress?)

One could counterargue that for any particular person, he'd rather take the
mediocre education, than flip a coin to get {great, or zero} education.

So it is at least conceivable that the society will be better off with
one course,
but any particular individual is better off with the other course.

There are certainly genuine examples of such contradictions. For example,
you personally are probably better off if you pollute, or if you steal
valuable stuff.
But if everybody did that, society would be worse off than if nobody did.

[Eric Sanders]

Let's say "okay" is exactly between "great" and "poor."

What do you think? Tie?

[Dale Sheldon-Hess]

On Thu, May 9, 2013 at 11:17 AM, Eric Sanders <er...@electology.org> wrote:

Let's say "okay" is exactly between "great" and "poor."

What do you think? Tie?

Yup.

Although Warren brings up a great point about how society's utility is quite likely non-linear here, and that it benefits much more from half the population having great educations.

--

Dale

[Clay Shentrup]

It depends on what you mean by great or poor. I want the outcome that maximizes the net utility. This is the only rational choice.

ScoreVoting.net/UtilFoundns.html

[Eric Sanders]

In the first scenario: Let's say 1/2 the students score their education as a 0 (out of 10) and the other 1/2 of the students score their education a 10.

In the second scenario: Let's say all the students score their education as a 5.

Do you view these outcomes as equally beneficial?

[Stephen Unger]

Why should we argue about such an artificial, overly constrained,
choice? In a decent society, everybody, regardless of their wealth,
ought to have the opportunity to develop fully their intellectual,
artistic, and perhaps, physical potentials. There should be no reason
for the kind of trade-off implied by the question. The only arguments
might be about definitions of the potentials to be developed.

The Nordic countries (as one example) are close to realizing this
ideal. The current situation in the US is disgraceful. We are steadily
regressing in this area. Around 1920, my father, son of a poor,
widowed, mother, living on the the lower East side of Manhattan, after
an excellent public school eduction, was able to get a first college
class education via evening classes at the City College of NY, with no
tuition charges. Currently, altho US income per capita is vastly
greater, this would not be possible.

Why this is so might be a better discussion topic.

Steve
............

Stephen H. Unger
Professor Emeritus
Computer Science and Electrical Engineering
Columbia University
............

> Is there a formula that could help us decide which is more *just*?
>
> --
> You received this message because you are subscribed to the Google Groups "The Center for Election Science" group.
> To unsubscribe from this group and stop receiving emails from it, send an email to electionscien...@googlegroups.com.
> For more options, visit https://groups.google.com/groups/opt_out.
>
>
>

[Dale Sheldon-Hess]

On Thu, May 9, 2013 at 12:22 PM, Eric Sanders <er...@electology.org> wrote:
> In the first scenario: Let's say 1/2 the students score their education as a
> 0 (out of 10) and the other 1/2 of the students score their education a 10.
>
> In the second scenario: Let's say all the students score their education as
> a 5.
>
> Do you view these outcomes as equally beneficial?

If you're going to call yourself a utilitarian, you pretty much have
to. But this setup isn't going to tell you much that's useful for the
real world, or likely to be informative for a utilitarian choosing
education policies to support.

- Contra the example, it is unlikely that all students will see equal
utility for all education levels, in which case matching students to
the right level for them will be important.
- There are non-student members of society; what are their utilities?
- The utilities for great/okay/poor education could vary widely based
on how many other students get them; someone's got to do the menial
work, so the 99th percentile going from okay to great education
probably isn't going to result in better careers (or relatedly, more
utility) for them. (Consider that some claim that the Arab Spring was,
in part, driven by an overabundance of better-educated students
finding themselves with no work that matched their skills.)

Given the information you've presented, your two scenarios are equally
good. But the scenario you've presented, other than sharing the words
"student" and "education" has very little to do with decisions about
student education.

--
Dale

[Abd ul-Rahman Lomax]

At 02:03 PM 5/9/2013, Warren D Smith wrote:
> >> Which would you prefer:
> >>
> >> A great education for 1/2 the students and a poor education for the other
> >> 1/2?
> >> OR
> >> An okay education for all the students?
> >>
> >> Is there a formula that could help us decide which is more just?
>
>--interesting.

Yeah, interesting question with responses that say more about those
commenting than about society and utilitarianism.

Basically, ask a question which is unanswerable without specifying a
universe of assumptions, and the answers you get tell you about the
people answering, not about the supposed topic of the question. Which
in this case is extremely poorly defined.

Is education some critical survival condition? Are people happier
with an education or not? Is happiness a value, and is it correlated
with "great education"?

And what is this "justice' thing, anyway, and how do we measure it so
we can make comparisons?

>One could argue that (maybe) society is going to be way worse if NOBODY gets a
>really great education. (No scientific progress?)

Yes, it could be argued. There are an infinite number of accessory
conditions that could be asserted as possible.

I consider the value of scientific progress to still be
controversial. Yeah, there are some great benefits. But ... there are
also some costs.

Totally unconsidered in the question is cost. I could *guess* that
the question has set up a zero-sum game. There is a fixed amount of
this thing called "education," and it is to be divided up. So, should
everyone get an equal amount or should it be divided in some other way?

Does everyone *want* education? What if one student wants to sit in
the sun and enjoy the weather, and another wants to dive into the books?

"Justice," however is a concept with heavy moral implications.
Injustice is "bad." What does that mean? What does "bad" mean? Whose
idea of "bad"?
"Bad" isn't something discernable in reality, it's something we
create through interpretation. It's a story.

It's *very* human to create these stories. But that doesn't mean we
need to believe that they are merely something discovered in reality.

>One could counterargue that for any particular person, he'd rather take the
>mediocre education, than flip a coin to get {great, or zero} education.

Or lift a finger. Or spend a dime or the time.

Or ... education is inequitably made available, with this or that
group disadvantaged and oppressed. Just more story.

There is no cheese down this hole.

Education is valuable, my opinion, my choice. However, not all
education is equal. Some is cheap and extremely valuable, some is
expensive and next to useless.

Useless for what? That returns us to the fundamental question, the
source and definition of "value." I've pointed to happiness as some
kind of metric. Maybe we could make a happiness meter, it's not
impossible. Would we agree on what it indicates? How would you feel
about someone else telling you how happy you are?

And that hints at something else. We create happiness ourselves,
through what we declare. Very much, we value that freedom.

>So it is at least conceivable that the society will be better off with
>one course,
>but any particular individual is better off with the other course.
>
>There are certainly genuine examples of such contradictions. For example,
>you personally are probably better off if you pollute, or if you steal
>valuable stuff.
>But if everybody did that, society would be worse off than if nobody did.

Only a narrow and befogged mind would conclude that one is better off
by being a thief. While there are exceptions, to be sure, people
steal because they believe they are forced to. They are not happy
people. A normal human is a social animal; our own happiness is
linked to that of everyone around us.

Range voting works with certain assumptions about the relationship of
range votes with personal assessment of utility. It works with
*simulations* that use simplified models of utility and
"utilitarianism" that makes assmptions about utility summation over
the electorate. It is not necessary to conclude or believe in
"utilitarianism," that maximizing the raw sum is an absolute good,
and that *distribution* or *variance* is irrelevant.

If we set up a situation where a naive summation analysis allows
serious harm to a few, while spreading out some good to many, to
maximize the sum, we run the risk that the few who are seriously
harmed will then have high motivation to *respond*. One person can do
a great deal of damage, if they are sufficiently motivated. So there
is something to be balanced against simple summation.

The process starts with the concept of majority *consent,* but that's
inadequate. The *goal* should be, for a true social welfare
optimization, unanimous consent. In any given situation, that goal
may not be attainable, but it's worth an effort to approach it.

[Abd ul-Rahman Lomax]

At 03:16 PM 5/9/2013, Clay Shentrup wrote:
>It depends on what you mean by great or poor. I want the outcome
>that maximizes the net utility. This is the only rational choice.

Well, it's insufficiently specified to be the "only rational choice."

"net utility" is unspecified. It's only a *concept* unless it can be
calculated from objective measures. We can use subjective measures,
but ... how, then, do we sum them?

We can perform the summation in simulations that *assume* absolute
utilities. The classic objection to utility summation in voting was
the lack of *actual commensurability.* However, the simulations show
that Range Voting works quite well if people have summable utilities
that can be compared with voting results. However, we are making,
then, an *assumption* that this extrapolates well to situations where
we don't know such utilities.

I think that's a reasonable assumption. If a voting system works well
when people have some sort of objective, summable utility, then it is
a *reasoanable* assumption that it will also work in other
circumstances. But it's going further to consider this the "only
rational choice." That's a conversation-stopper. It indicates a
closed mind, or at least one give to hyperbole.

*If net utility is completely defined,* I agree. But such utility
will include *all measures*. I.e., suppose there is a utility to
"equity," and "fairness," and lack of extreme consequences to individuals?

If *all measures are included,* yes. But the example was a narrow
question, with practically no information about details that would
matter very much.

"Net utility" is a great concept, and it's actually *useful.*
However, in actual practice, net utility is often the very point in contention.

[Abd ul-Rahman Lomax]

At 03:22 PM 5/9/2013, Eric Sanders wrote:
>In the first scenario: Let's say 1/2 the students score their
>education as a 0 (out of 10) and the other 1/2 of the students score
>their education a 10.
>
>In the second scenario: Let's say all the students score their
>education as a 5.
>
>Do you view these outcomes as equally beneficial?

To whom? Who is making the choice here, and on what basis?

Are we only concerned about subjective self-assessment?

A more classic measure of educational effectiveness has been the
comparison of the income of those with higher levels of education and
those with lower levels. But individuals vary *drastically* on these
measures. But at least this approach can study cost-effectiveness. It
can also consider return on social investment in subsidized education
vs. increased tax revenue.

Completely neglected is any concern for the opinions of those who are
paying for the education.

Now, if I were a school studying, with this poll, the effectiveness
of our education, I'd be very concerned about both results. They are
equally problematic, but for different reasons. With the first
result, we are in abject failure with half our students! Look, we
might be able to get this result by picking our students by lot and
only spending on half of them. Or by their last name compared to
alumni, say....

The second result would worry me because it is totally unrealistic,
and thus calls the poll into question. It actually means nothing
other than that. If I've invested years in education, and I rate it
as a 5, well, I think I'd be awfully disappointed!

As an educator, I'd be aiming for *all students* to be at a 10. As an
educational manager, and if an employee of the school was content
with either of these outcomes, I'd be tempted to fire them. At least
they need some training in how to stand for excellence.

This problem is the kind of issue that is brought up by people trying
to make mincemeat out of utilitarianism, typically. Eric is free to
disagree, but ...

utilitarianism is only a *tool*. Range Voting is "rational
utilitarianism," technically. Tools are either useful or not, if we
attempt to categorize tools as true or false, we get into
irreconcilable arguments.

Is Range voting useful? Obviously, at least under some circumstances,
Yes, I have no doubt about that.
Is Range voting "true?" Hello? Are oranges 27?