Exodus 22:25, Leviticus 25:36, and Deuteronomy 23:20-21 forbid Jews from charging interest on loans to "your brother" (other Jews).  (This is to me the most convincing argument against Judaism and Christianity, because it's too simple to argue around.  That proscription is just wrong, in exactly the way you would expect laws written by uneducated tribal people to be wrong.)

Roman Catholics believe they must follow the Old Testament laws, except for the ones they don't have to follow; but during much of the middle ages in Western Europe, this was one of the ones they had to follow.  They interpreted "your brother" as meaning "brother Christians".  So Jews could lend to Christians with interest (and, presumably, Christians could lend to Jews).  This was convenient for everyone.  The Jews were necessary to work around an irrational moral prohibition of the Christians.

Of course, the Jews had to take on the guilt of violating the moral code, even though it was for the benefit of the Christians.  (This was also convenient; it meant that after some Jews had loaned you an especially large amount of money, you could kill or expel them instead of paying them back, as the Spanish monarchy did in 1492).

Later on, some orthodox Jews hired goyim to turn lightswitches and other electric devices on and off for them on the Sabbath.  They're called Shabbos goy, the Sabbath goy (thanks, Alicorn!).

JCVI is considering moving from an on-site hardware grid, to cloud computing.  There are lots of reasons to do this.  One is so that Amazon can be our Shabbos goy.

We develop lots of bioinformatics software that we're supposed to, and would like to, give out to anyone who wants it.  But if you don't have 800 computers at home, connected using the Sun Grid Engine with a VICS interface and using a Sybase database, with exactly the same versions of C++ and Perl and every C++ and Perl library that we do, you're going to have a hard time running the software.

We can't put up a web service and let anybody send their jobs to our computers, because then some professor is going to say to their freshman class of 200 students, "Today, class, your assignment is to assemble a genome using JCVI's free genome assembly web service."

If we could charge users just a little bit of money, just a fraction of the cost of running their programs, we could probably do this.  Then people wouldn't be so cavalier about running a program repeatedly that takes 500 CPU hours each time you run it.

But we can't, because we're an academic institution.  So that would be evil.

So we need a Shabbos goy.  That's Amazon.  We can release our software and tell users, "All you have to do to run this is to get an account on the Amazon cloud and run it there.  Of course, they'll charge you for it.  They're evil."

(The Amazon cloud is evil, BTW.  They charged me for 21G of RAM and then only gave me 12, and charged me for 24 1GHz processors and gave me about 1/4 of that.  I spent over $100 and was never able to run my program; and they told me to stuff it when I complained.  But that's another story.)


  • People would rather patch around failings in their reasoning than fix them.
  • If your morals require you to use the services of someone not adhering to your morals, you may be in error.
P.S. - My morals may require me to use the services of someone not adhering to my morals.  I believe in a moral ecosystem:  You don't hold your dog to your moral standard; and you don't remedy this by adopting your dog's moral standard.  But AFAIK I'm the only one.  People who believe in one universal moral code shouldn't use Shabbos goyim.  (I don't think that orthodox Jews believe gentiles are supposed to obey the Torah, so their use of Shabbos goyim may be logically consistent.)
New Comment
92 comments, sorted by Click to highlight new comments since:
Some comments are truncated due to high volume. (⌘F to expand all)Change truncation settings

Presumably they also went to non-Jewish doctors if they had a medical emergency on the Sabbath.

Nitpick: Jewish law permits and in fact requires the breaking of Biblical commandments if someone's life is in danger (exceptions: idolatry, murder and breaking sexual taboos). In fact, there's an old Jewish joke that I think gets at some of the moral weirdness you're bringing up here:

A Jew and his Christian friend are sitting down for dinner at a restaurant. The Christian recommends the cheeseburger, but the Jew says that, since he keeps kosher, he's forbidden from eating any meal with both meat and dairy.

The Christian asks whether the Jew would eat a cheeseburger if his life depended on it, and the Jew says that Jewish law allows someone to break a ritual law if zir life is in danger. So the Christian draws a handgun, points it at his Jewish friend, and says "Order the cheeseburger, or I'll shoot!"

After dinner, the Christian puts his gun away and apologizes to the Jew. "You know I never would have shot you," he says. "I just wanted to see what you would do. Please forgive me."

"Forgive you?!" asks the Jew. "Of course I won't forgive you! Why didn't you make me order the bacon burger?"

In conclusion, if your morals make you wish someone would come along at gunpoint and force you to break your moral system, you may also be in error :)

Everyone's favorite quotation from the Confessions of St. Augustine: "Grant me chastity and continence ... but not yet."

using terminology from my goodhart's law post, In most religions, G is to be a good person and G is the bunch of laws to be followed. The presence of Shabbos Goyim anywhere in your process is a very serious case that your Gs are flawed.

Nice Post and Nice discussions.

For those not intimately familiar with past writings in this community, the post On Goodhart's Law offers a useful terminology for talking about formal goal oriented processes and a number of top rated comments with relevant links and insights.

If we could charge users just a little bit of money, just a fraction of the cost of running their programs, we could probably do this. Then people wouldn't be so cavalier about running a program repeatedly that takes 500 CPU hours each time you run it.

But we can't, because we're an academic institution. So that would be evil.

You mean like the academic journals that charge subscription fees?

People wouldn't have a problem with an academic journal that they believed charged a reasonable fee. But consider a typical journal published by Springer-Verlag or Mary Thomas Liebert: * Subscription fee: $100 per issue, $1200/year * Cost per article for non-subscribers: $30 * Fee paid by the author of each article: about $3000 * Advertising fee: I'd guess at least $5,000/page, based on the fact that each full-color page in your article usually adds $1000-$2000 to its cost * Salary paid to reviewers: $0 * Cost of electronic publishing: about $100 per issue, divided among all subscribers They have a problem with having a small number of subscribers. But many hobbyist groups manage to publish quality journals to equally-small audiences at a cost of under $10/issue. The fact that people aren't jumping in to compete with lower-costs journals makes me suspect that it isn't that easy. But it's still not at all obvious why academic journals cost so much. (The big ones, Science and Nature, are relatively inexpensive.))

But it's still not at all obvious why academic journals cost so much.

Sure it is. Subscriptions are mostly paid for by institutions, rather than individuals. Any given article effectively has a monopoly on it's own content, so once a university has subscribed and the profs are used to getting free access to that content, it's politically difficult for the university to un-subscribe. Then the journal incrementally increases the prices. Soon, the university's accounting department is on the losing end of a frog-boil.

I'm stealing this expression.
I'm going to use "frog-boiler" as an insult.
The more widely-accepted term for nondestructive appropriation of creative content is 'piracy.'
Vocabulary is not creative content.
Vocabulary absolutely can be creative content -- boiling something down to a few words is a difficult art. I would agree with it not being covered by various intellectual property laws.
-- Peter da Silva
We have no disagreement here.
It's not one I accept. I think it's a very bad analogy, and refuse to use the word with that meaning.
How about, I'll use it fairly with attribution where possible
Are you going to let that frog-boiler intimidate you?
I just want to call someone a frog-boiler.
Apparently I gave the wrong impression that I was arguing that academic journals are evil. I actually meant to challenge the idea that an academic institution charging for its services is evil, using academic journals as an example of institutions that currently do so and get away with it. Other examples include text book publishing and undergraduate tuition.
I assumed that was what you meant; so I explained why I still think academic journals are suspect. I don't think either of us has a definition of "evil" that would support argument on this topic anyway.
Huh? People are most certainly jumping in with zero-cost (to read) journals such as PLoS and others. The open-access publishing movement is not obscure and I'm surprise to see that people here aren't aware of them. The reason existing journals cost so much is that publishers can charge monopoly rents based on their ownership of a high-status imprint. That game is not going to last very much longer, IMO.
PLoS is a non-profit, and I'm certainly aware of it. If, however, for-profit academic journals charge much more than it costs to produce them, I would expect to see for-profit startups competing with them. The "impact factor" measure is a part of this; you can't just start up a new journal and have a high impact factor.
PLoS claims to be sustainable on internally generated revenue at this point. I was just at their conference.
Possible explanations for not seeing that: * There are for-profit startups competing with the big boys, but we just don't know what they are, because of course they're smaller and less visible than the big publishers * Startups get bought out by big publishers * Social barriers to entry like the impact factor issue you mention * Social barriers to entry amplified even further because startups are a bunch of entrepreneurs breaking into the field from outside, so they don't have the same connections as scientists working in the field (less of a problem for big publishers because they can buy existing publishers)

That proscription is just wrong, in exactly the way you would expect laws written by uneducated tribal people to be wrong.

There's nothing wrong with assuming that in the ancient past people were less educated than they are now, and the general claim "religious beliefs are stupid" has been proven to the LW community's satisfaction. You, though, are making two claims that go far beyond these truisms: (1) that uneducated tribes are stupid in some specifiable pattern X, and (2) that "don't lend interest" exemplifies pattern X.

I wish you... (read more)

I don't see what you are disagreeing with. The proscription is wrong today, however useful it was then, because it was not written by people aware of how things would change in the time since its creation. A god with the power to predict the future could have created a better code, but human beings who didn't even know what capital was couldn't be expected to (and, in fact, didn't). Your remarks are apt and informative (I upvoted), but PhilGoetz hasn't actually written anything incorrect, yet. I approve of putting a go stone out to block the common error, but I wouldn't have worded it as a disagreement.
Well, thanks for your compliments and your upvote. I don't say openly that I disagree with PhilGoetz, but I guess to the extent that my disagreement shines through anyway, it's because I think the particular law he's chosen is only very weak evidence against religion, whereas he claims it's very strong evidence. Granted, an omnipotent, omniscient God could certainly have done far more useful things for the Bronze Age peoples of the Middle East than simply advising "Don't charge interest;" I like some of sci-fi author David Brin's remarks along those lines: If a deity were to suggest a ban on commercial interest, though, such a suggestion would hardly be worthless, let alone incorrect, as PhilGoetz appears to claim. A minimally powerful god who could predict the future might have bothered to warn that commercial lending would be required in the future, or she might have figured that it was too much hassle to explain why and when it would be required to a tribe with little concept of scientific history or economics. There is no evidence in the text of the Old Testament that the law against commercial interest was meant to be perpetual; Jewish legislators could and did change it as the basis of the economy changed.
Fair enough! And a great Brin quote (edit: found the link!), too - and on a similar subject, though I wouldn't normally recommend this book: one of the things I found interesting about Sheri S. Tepper's novel The Fresco was that it rather undermines several kinds of theodicy by having not-all-that-powerful aliens intervening in a number of quite huge terrestrial problems.

In regards to your postscript, it's certainly possible that the use of Shabos goyim by Orthodox Jews is logically consistent, but it doesn't stop there! One of, to me, the most fascinating examples of this sort of patching is Shabbat technology. The example that always strikes me is the Shabbat elevator, which is quite common in Israel. Basically, it's a normal elevator, except that during the appointed period of time it operates in a much different manner: constantly cycling through the floors and stopping on every single one. This way 'observant' Jews ar... (read more)

I'm not sure why you picked hasidic. It sounds like you are confuing hasidic with haredi which is roughly speaking the general ultra-orthodox population. The hasidim are a specific strain of orthodox Judaism which arose in the last 1700s. But many haredim are not hasidim. In any case, the Shabbat goy has nothing to do with either the larger category or the smaller category. Note also that neither category describes all of Orthodox Judaism.

You are incidentally correct that orthodox Judaism does not believ that gentiles have to obey the Torah. However, I'm ... (read more)

Wikipedia says, "Haredi Judaism comprises a diversity of spiritual and cultural orientations, generally divided into Hasidic and Lithuanian-Yeshiva streams from Eastern Europe, and Oriental Sephardic Haredim." I wasn't aware. Changed to just say "some orthodox".

I've talked with an Orthodox Jew who thought the Shabbos goy was very contrary to the spirit of the law-- he said (I haven't verified this) that Shabbos goyim were first instituted in north-eastern Europe where keeping a fire going in the winter was a matter of life and death, and therefore required-- but that Shabbos goyim were eventually employed for matters of mere comfort and convenience.

It's not just against the "spirit of the law", it is against the letter of the law in Orthodox Judaism.

The Amazon example doesn't seem to be that illustrative of the concept you are trying to get across, mostly because the reason academic institutions don't sell computation is that they aren't set up for it, not that commerce is considered evil. They have no problem charging for other services, such as tuition.

Here's a better one: police, military, and government in general. Everyone in that role has slightly different moral codes than the rest of us, in that they are able to legitimately employ violence in various forms, and for the most part we are willing to cede that role to them. The government is our shabbos goy, although too often a master rather than servant.

Commerce is considered evil. We're sensitive to this at JCVI, because a large proportion of the biology research community hates our founder, Craig Venter, for having had the gall to make money off of biology. Today, it's better-accepted; but when he started, it was apostasy. If he hadn't made money, there wouldn't be a JCVI now; we couldn't afford to do what we do just on grants. By its fruits you shall know the tree. Universities are given a free pass in some ways. Note that the first universities (Padua, Paris) were deliberately for profit. Harvard would make a profit without charging any tuition - so why does it charge so much tuition? MIT may be in the same situation; they get over a billion dollars a year from the government, and have only 10,000 students. Many of our grants require us to give the results away for free. If we charged people to run the software on our servers, we'd have to make the case that charging for the use of computer time was distinct from charging for the use of the software resulting from the contract. There's little legal precedent in this area AFAIK. Moreover, legal precedent don't matter when you don't get a chance to make your case - it doesn't matter what the law says; it matters how the next program officer feels about what you're doing.
Thanks for the illustration - I was wondering about the details.

I find it fascinating that the original post's "We need a jew" was unacceptable, but the modified post of "We need a Shabbos goy" is perfectly fine. Especially because in the original post the moral failing was the christians', whereas in the modified post the moral failing is the jews'.

And even more so because I too feel emotionally motivated to dislike and down-vote the first while supporting the second, specifically because the first one feels anti-semitic and the second doesn't... even though the opposite seems to be true.

Perhaps it's because the term "we need a jew" signals ownership, resulting in objectification and degradation of the subject? We need a Robin Hanson in here!

I wonder if the low rating of this post is because people are more touchy about Jews than they are about Christians. People here think it's fine to criticize Christianity all day. Shouldn't it be equally acceptable to criticize Judaism? (I criticized both.)

I'm going to put up 4 replies to this comment, in matched pairs for karma balance. If you vote, vote on both comments in a pair. Comment if you found one pair harder to vote on.

ADDED: Wow, the results surprised me. I would have found it hard to upvote "Judaism is stupid". My brain would direct it, but my finger would protest.

Christianity is stupid.

Judaism is stupid.

You are probably right. The kind of thinking that implements political correctness does not seem to be particularly compatible with the abstract thought involved in making their disapproval consistent. For what it is worth I like the post for the content but probably would have enjoyed it slightly more with the original title. I just love provocative irony.
I just wasn't that interested in the subject matter (didn't vote up or down) but you do use the word "Jew" a lot. And I'm a little uncomfortable with it as a general term for a way to work around ethical rules. I understand the point you're making and I'm not offended but the last couple thousand years of antisemitism might distract from your point.

I see - If people use "Jew" to mean "the agent you need to help you work around problems in your morals", the negative association bleeds over onto the word "Jew", even though it's the Christian (in the money-lending example) who's committing the error.

Writing is more memorable when you use emotionally-laden terms, but also riskier.

I used the term "Christian" a lot, too; and implied the Christians were the ones at fault. But that wasn't uncomfortable?

I'm thinking how to reword it so that it doesn't go flat, but so far nothing comes to mind.

Is there an existing term to refer to the person you use to commit your sins for you? Scapegoat isn't right. Is there a word for a person hired to operate electric devices for you on the Sabbath?

In looking for the answer to this, I found this absolutely wonderful page explaining different rules about the use of electricity on the Sabbath. Studying it would make an excellent preparation for law school. Here's a quote:

It is forbidden on the Sabbath to set the alarm of a battery-operated transistorized alarm clock or a wind-up alarm clock even though the alarm was wound before the Sabbath. The reasons

... (read more)

Is there a word for a person hired to operate electric devices for you on the Sabbath?

Shabbos goy.

Awesome! So it's less-offensive to use the term "Shabbos goy", which logically criticizes Judaism, than to use the term "Jew", which creates a negative association with Judaism.
If that's so, then if you tell people what they hear are the associations
Got here late -- what was the original title?

"Kill the Jews".

I guess some people are just touchy.

AHAHA! Well done.
Necessary Jews

When I first saw it under that title, before I read the article, I read it as contrasting with "Sufficient Jews" or possibly "Contingent Jews".

I'm still enjoying imagining what those could possibly mean.

Jew is just a really loaded word because it is used by various sorts of supremacists. "Jewish person" can be substituted without any kind of negative connotation.
Nah. Your word choice is now orthogonal to the usual patterns of antisemitism and you have legitimate cause for making fun.
Is the plural Shabbos goyim?
I'd assume so. I'm only half-Jewish on my dad's side, and he was brought up in a fairly liberal sect and ditched even that shortly after getting bar mitzvahed and didn't pass on a Jewish surname - I'm as non-Jewish as one can get while retaining any claim to it at all. So I'm not a proper authority on the subject.
I thought that you're considered Jewish if your mother was Jewish, and not Jewish if she wasn't.
That depends on the sect, apparently.
Exactly. Genetics, of course, don't give a damn whence the Jewishness, and it didn't inhibit my ability to pick up secondhand culture either except inasmuch as that set of grandparents died when I was a kid, but my point was that it being on my dad's side makes me "less Jewish" by a relevant metric.
Jack is right that it's distracting, and Phil is right that it shouldn't be.
"Henchman" or "cat's-paw" seem fine to me.
Christianity is stupid.

This is a nice and unexpected parallel. I think I may notice more instances now that I know the general pattern. Thanks for writing it up.

The orthodox jewish position is consistent with using shabbos goyim. They maintain that the jewish nation was created for a specific purpose ("light to the nations", whatever that means) and their legal (vs. moral) code is the result of this setup; it is only meant to apply to the jews. There is a separate body of law "seven laws of children of Noah" that are universal, and much closer to a "moral code". http://en.wikipedia.org/wiki/Seven_Laws_of_Noah

My funny story, though cannot confirm that it's true, was about a very capable shabbos goy who after some years turned out to be jewish.

I don't know if you posted this before I wrote the PS to my post; but I agree that it's possible to call this consistent. (And, oops, I meant orthodox, not conservative.) I think, though, that saying "These are the rules given us by God by which we show our dedication to Him", and then finding sneaky technical ways around them, like setting a timer to turn your television on at 8PM the next day, would not impress the Abrahamic God as much as it would impress the American shyster-lawyerish Old Scratch.
Yes, I misunderstood the conservative reference. As far as the hair-splitting, there are a couple of ways to view it. On one hand, Talmud has a concept of "being despicable (person) with torah's permission", meaning technically following the law but not its intent. This refers to cases where some moral violation is involved, such as stealing, or, as in your example, defeating the purpose of Shabos (which would be reflecting on creation) by setting up the tv timer. In other areas, where the law has no obvious moral intent finding a hack is considered praiseworthy. So why have these amoral laws in the first place? Maimonides held that besides formalizing the moral principles (how to treat stealing in a system of courts) the law is to train the mind, which makes you think through a lot of things you do throughout the day. Orthodox judaism has a lot of reasonable answers for a lot of things. This is why I found it so damn hard to abandon, it's so surprisingly consistent in many places. I think there is something to be said of meme survival that causes religions to become close to reasonable over time, and judaism probably had the longest run out of all of them.
David Friedman has just posted on this. As it turns out, the debate about finding "creative" loopholes of Biblical law is isomorphic to the controversy over constitutional interpretation in the U.S.
Is there a rabbinical school analogous to the "living constitution" school?
The left wing of the Jewish Committee on Law and Standards, which is in turn a part of the American sect called Conservative Judaism, comes close. http://en.wikipedia.org/wiki/Conservative_Halakha#Difference_in_methodology_from_Orthodoxy http://en.wikipedia.org/wiki/Committee_on_Jewish_Law_and_Standards#Process
Funny story please
It's slip on banana kind of funny. These people were very religious and rather disappointed to find out they have been sinning for quite a few years.

Relevance to LW?

If your morals require you to use the services of someone not adhering to your morals, you may be in error.

It's also interesting to see how people patch around failings in their reasoning instead of fixing them.

I suggest adding this comment to the main article, as it provides an excellent summary.

Interesting, but I don't understand how Amazon helps you. If you release your software in a form that users can easily upload and run on EC2, presumably you'll have to simplify the dependencies anyway (Sybase, huh), so users can just as easily run it at home. Or am I missing some part of the story?

No, you can package it as EC2 image - which has identical hardware (virtual), identical software (whatever you put there) etc. People just clone your image, and start it. It feels very much like starting a new lifeform from genetic instructions from pristine zero state vs trying to genetically engineer existing live organism (a much more difficult thing).
Thanks! This manner of "releasing" software counts as evil in my book. It certainly goes against the spirit of open source. It's just like releasing code that depends on an expensive Oracle license.
I don't see anything evil about it. Are Knoppix-based live cds evil? They're very similar things. It's usually just a standard image of two year old CentOS (or whatever they decided to use - for some reason CentOS seems most popular, even though I'd much rather have Ubuntu everywhere) + a bunch of tarballs + some initialization scripts which unpack and sets it all up. Very little about EC2 is really EC2-specific (it's a different matter about S3 and some other Amazon services), so if you have a spare old x86 Linux server and are willing to suffer from exactly the same distribution they used, you'll be able to get it running in little time even without terribly much clue. If you want to use different architecture / distribution / or even much more recent CentOS than the image - well, that will require some clue and some effort - but it would anyway, and at least this way people releasing the software don't have to support it. The idea of write-once-run-everywhere software is extremely far removed from reality. Even write-once-run-on-every-x86-Linux just doesn't work in practice. The cause of problems are distros which never got together and set up a single standard, not people writing software that just works on a single distro.
If we run our software here using the same front end, we don't have to modify it to make a release. That is, we wouldn't write it to assume use of the Sun grid engine and Sybase if we were using the cloud. (I hope.) Yes, if we said, "Here's our software, which will run on the cloud provided you have a Sybase license," that wouldn't be a huge improvement. Also, users usually can't run these programs at home, since they require dozens or hundreds of computers.

If we could charge users just a little bit of money [...] But we can't, because we're an academic institution. So that would be evil.

I doubt very much that the people who made this decision were considering the moral implications. Far more likely, they decided based on economics. It sounds like a good decision to me, from what you've presented here.

I really don't think it's appropriate to equate a business decision with religious craziness, not without more evidence that they're behaving irrationally.

I didn't say the decision was bad. (The decision hasn't been made yet, BTW.) Even the part about needing Amazon to get around the moral issue isn't a bad decision, since we have no control over academia's morals. Just pointing out another instance of a general phenomenon.

I had heard that Christians at some point were forbidden from charging interest on loans, and I had heard that Jews at some point were forbidden from charging interest on loans, but finding out that Jews were allowed to charge interest to Christians but not to other Jews and vice versa was a WTF moment for me.

The left wing of the Jewish Committee on Law and Standards, which is in turn a part of the American sect called Conservative Judaism, comes close. http://en.wikipedia.org/wiki/Conservative_Halakha#Difference_in_methodology_from_Orthodoxy http://en.wikipedia.org/wiki/ valentines day gifts for her Committee_on_Jewish_Law_and_Standards#Process

In Orthodox Judaism, based on Biblical passages, the Sabbath was given to Israel. As you mention non-Jew's are neither expected nor encouraged to keep the Sabbath.

That being said, despite the popular notion of a "Shabbos Goy" it is generally forbidden in Orthodox Judaism to benefit from a gentile's violation of the Sabbath. One is certainly forbidden to tell a non-Jew to break the Sabbath.You are not allowed to have a non-Jew turn on the light for you and he did it without being asked you may not utilize the light. While there are some excepti... (read more)


My experience with Amazon cloud has been rather positive. Very much unlike my experience with live sysadmins, who are all terrible human beings by nature of their job. I'd really rather not go off the cloud ever.

I don't think I'm a terrible human being, on or off the job.
I'm sorry, but you have to update that belief now. :)
I've heard the data point several times before in different contexts, so minimal updating required. As far as I can tell, it comes down to different incentives. End users have the goal structure G* (most closely aligned with the actual business), developers have Q*, sysadmins have D*, and none of them are well aligned with each other.
I hadn't heard it. I wasn't aware that some people were prejudiced against sysadmins.