Computer: Game: Simulation
Sim Utilitarianism   (+1, -2)  [vote for, against]
Illustrate flaws in consequentialism with a computer game

Utilitarianism, in case anyone doesn't know, is that actions are right in proportion to the promotion of the greatest happiness of the greatest number, and can be generalised to consequentialism - the end justifies the means.

[Eleventeenthly] was playing a game earlier which involved balancing piles of smiley faces, which turned into frowny faces if they were about to fall over. The score depended on whether the pile stayed upright or fell over.

Good game. Anyway, it occurred to me that it would be really interesting if the score depended instead on how smiley or frowny the faces were. This would be utilitarianism - the more and the smilier the faces are, the higher one scores.

So, i imagine a game a little like Conway's Life but with smiley/frowny faces instead of cells. The object of the game is to achieve the greatest happiness of the greatest number, and this can be both done and measured in various ways, illustrating the flaws in utilitarianism:

* Feeding and caring for the faces makes them happy. Friendships make them happy. Loneliness tends to make them sad, a couple of neighbours makes them happy but lots of neighbours makes them sad, though not as sad as loneliness. Therefore, overpopulation increases the total amount of happiness while reducing the mean happiness.

* Utility monsters: Some faces are able to be much happier than others, others much sadder and still others can experience more highs and lows. These are indicated by colour. They can be bred together to make children which inherit certain characteristics, perhaps unreliably, so they have a "genome" which determines their capacity for pleasure. The ones with the greatest capacity for happiness are also the most fertile. Therefore, it makes sense to breed happy faces and kill unhappy ones. However, killing unhappy ones can make their friends and relatives sad. That's OK though, because they can be ostracised or shunned before being killed.

* Injustice: Faces can be falsely or justifiably accused of crimes. They can be publically tortured or executed, which increases the pleasure of the spectators because they feel justice has been served even if it hasn't, and they also have bloodlust.

* Higher and lower pleasures: One way of making the faces happy is to play with them, feed them and so forth. Another way is to inject them with drugs and leave them to starve. There are also orgasmatrons and theatres. Orgasmatrons guarantee happiness, but theatres offer a fifty-fifty chance of an uplifting lighthearted play or a harrowing but improving social drama. However, the pleasure achievable through the orgasmatron is always more than the theatre. Faces can also be given the chance to have a "lobotomy" which turns them into happy zombies which can be enslaved, possibly by utility monsters, or they can escape into a life-long fantasy world.

The scoring system can be modified for modal, mean or median measures of happiness, and to take into account pleasure and suffering, so that different measures of happiness and unhappiness are available, plus separation of higher and lower pleasures, levels of freedom and so on.

Some of this i can see would be easy to implement because it's very like cellular automata or tamagotchi, but other bits might be trickier. I can't decide whether this should just be smiley and frowny faces in a "Life"-style grid, whether it should be like the Sims, or maybe like a role-playing game.

Another aspect of this is that it focusses on utilitarianism. I'm also interested in looking at other ethical theories through games, such as deontology, egoism, naturalistic moral theories or ethical scepticism, but don't know how. Maybe this game could be modified to do that too.
-- nineteenthly, Feb 25 2009

[+] I would play. You might enjoy a book I recently read, called "Growing Artificial Societies".
-- placid_turmoil, Feb 25 2009


There seems to be no consequence to the orgasmatron or lobotomies.

It seems to me the only flaws you're illustrating are the characteristics with which you have a problem. The thing largely comes across as a rant to me, sorry.
-- phoenix, Feb 25 2009


//It seems to me the only flaws you're illustrating are the characteristics with which you have a problem. The thing largely comes across as a rant to me, sorry.//
Not at all. I just ran through common objections people make to utilitarianism, copied some thought experiments and used them to illustrate how it might work. I don't feel very strongly about utilitarianism as such either way, because to me the contrast would be with deontology or some early twentieth-century metaethical theory, and i'm keener on virtue-based stuff so it's not very important to me at all.

[Bigsleep], there's Philosophy as in philosophy of life and philosophy as applied to isolated situations. It may be artificial to isolate something that way, but in the history of ideas, such isolation is important. Also, it's normal for people to isolate things in this way in other areas, for example with computers, although thinking of software and hardware together would be more holistic, it wouldn't be easy to hold all of that in mind at the same time. You could look at a substance abuse disorder physiologically, pharmacologically or socially. None of them have the whole picture but they can all make a contribution.
-- nineteenthly, Feb 25 2009


I don't know what Conway was trying to do, but it's an interesting question.
Incidentally, on the subject of advocacy, you might be able to come up with a weighting for higher versus lower pleasure which would help, and if the central tendency measure used was modal, it would also improve the accuracy of utilitarianism. That particular line of thought was part of one of my degree dissertations.
-- nineteenthly, Feb 25 2009


It's not connected to deontic logics. They try to map permission and obligation onto the modal operators of contingency and necessity. Your mention of Kripke, by the way, makes me nostalgic for an incident when [grayure] (wife) tried to show an interest in my stuff by reading a book by Kripke and gave up after about one sentence. Actually, i don't care so much about Kripke, though i do think you could find a way of applying logical theory to everyday life, for instance considering dissociative processes in personality and the metaphysical side of intellectual property rights, but it's not relevant to this and i don't know how to relate Kripke's stuff to that.

Anyway...

Deontic logic is to do with ideas of duty, and consequentialism in general, e.g. utilitarianism, is fundamentally opposed to the idea of duty. Whereas there might be a way of basing a game on deontic logic, i don't know how.

What i meant by modal was this. You could take the phrase "greatest happiness of the greatest number" in several different ways. You could see it as the mean happiness in a society, i.e. how happy everyone was divided by the population, but that would sanction a situation like doubling population while reducing happiness considerably, which is one of the big objections made to utilitarianism. Then there's median happiness - rank everyone's happiness from suicidal depression to ecstatic joy and find where the middle is. That's really hard to quantify. By modal, i meant the mode in statistical terms, so what i had in mind was, say you surveyed everyone and got them to rank their happiness on a scale of one to ten. The best society would then be one where the majority of people rank their happiness at ten.

Two obvious flaws in all of this are how to compare and rank happiness, so the whole thing is a bit silly with respect to that particular idea, but that doesn't mean other values can't be ranked. For instance, it might be valid, though probably wrong, to say a good society was one where most people were wealthy. That is at least measurable, though probably not very useful.
-- nineteenthly, Feb 25 2009


The fact that a big objection can be made to something doesn't necessarily mean it's wrong. However, regardless of the subject, each measure of central tendency has disadvantages. The mean is easily thrown out by outliers. I found this interesting when i noticed that a depressive and apparently statistically naive person became despondent when she looked at scores on the mean of an assessment done by a group of people on her. In fact, the results had been totally skewed by a couple of outliers, both of whom were the same person, and the median and mode were both a lot higher. In terms of utilitarianism, this would amount to the utility monster. Outliers really screw up the mean sometimes. The thing about modes is that they work well for nominal data, for example the mode for letters in English is "E", but there's no mean letter. There can also be more than one mode, so there might be a lot of happy and a lot of unhappy people, which wouldn't be as good as measuring the mean, which would then be in the middle.

When i was more into utilitarianism, i preferred the least unhappiness interpretation. This probably reflects my own depressive tendency. The GHP business links in with my whole difficulty in "getting" economics - what's included in the concept of economic value? Though i do broadly agree with what you're saying here, i'd question what you mean by "use". One possible use of "useless junk" is to transcend it and realise the emptiness of it all. Also, right now as i type this, my back is being warmed by someone else's "useless" junk which is providing me with heat, light and energy for cooking and drying clothes. I'm also sitting on a chair, typing on a keyboard and wearing a hoodie all of which someone else threw out as junk. Not so useless to me.

Chemistry has been, and presumably still is, an incomplete model. There was a time when noble gases were inert gases, and a time before that when phlogiston was a substance and salt an element. Nowadays, and i must admit my knowledge of chemistry is rather hazy, it's all about orbitals and, i think, ionic and covalent bonding blur into each other, or something. It's not complete now and it never will be. Likewise, alchemy was able to engage with the æsthetics and symbolism of materials in a way which is compromised in chemistry.

Sorry, i've completely lost track of what i was talking about!
-- nineteenthly, Feb 25 2009


OK, to try again without rambling!

The mean, median and mode all have pros and cons in various areas and mathematics is no exception.

Gross National Happiness is something i don't know enough about to comment on. I have assumed it attempts to include environmental and other values.

In sciences based on observation, as opposed to, for example, mathematics, and even that's questionable, there are working models and contradictions or unexamined areas. There is never a complete model. If there were, it would be closed off to new enquiry, an objection which was raised recently here on the HB to homœopathy - a closed, dogmatic system. That is alleged to happen in science too, according to Thomas Kuhn for example.

Alchemists had other motives for what they were doing. It was akin to a spiritual path rather than a science, and its aim was different. It could be claimed that economics is similar (but there might be another way, blah blah, got the T-shirt). Concerning software, a subject where i feel the ball got taken away from me when object-oriented programming took off, i can't comment, unless you're talking about the Open Source issue. All i have from understanding how software works nowadays is that there's a possible link between Quine's logical theory and the idea of objects. After that, it all becomes utterly obscure. I can't help thinking that if it didn't, i wouldn't now be breaking up furniture and burning it, but such is life.
-- nineteenthly, Feb 25 2009


Absolutely, in itself a huge pile of slightly computers, mobiles and the like is a deeply disturbing thing. They are not, however, utterly useless because knowledge is power. There is a way of using things even if "they" don't want you to. I know someone who tiled his shower with old CPUs. www.instructables.org for that sort of thing though. I would prefer the value of transcendence to that kind of use, but i still like the DIY approach to the "things that the everyday folks leave behind".

// That analogy was only a device to point out what gross incompleteness is // Sorry, i have a strong tendency to get bogged down in details (and here we are again). Concerning programming, well, eyes glaze over again but the words "creeping featurism" occur to me in that context. What is the need exactly? Fixing bugs and security holes, i presume, but i've generally thought that less must surely be more (says this avid user of emacs). Meaning that if there's less code, is there not less to go wrong?
Actually, tell you what. I can turn the problem into the solution here by attempting to bake this. I have written a version of Life from scratch because that's the kind of amateurish thing i can manage with programming. I can try to learn programming by baking this idea. If i attempt to come up with a more complete, formal description of this idea, find out how to implement that and write the bloody thing, i will understand the issues and have produced an educational tool.
-- nineteenthly, Feb 25 2009


Once again, i think this should be taken to email. Having said that, is it not the case that the more code there is, the more bugs there are?
-- nineteenthly, Feb 25 2009



random, halfbakery