Where meritocracy fails | Hacker News Reader

Where meritocracy fails

chesnok.com

78 pointsdraegtun44 comments

Is 'merit' necessarily unrelated to 'privilege'?

Much of our talent and ability is not something we are born with; a meritocracy based on innate ability would be a mess- it would practically boil down to ranking everyone based on their IQ. Instead, a meritocracy is typically going to be based on current ability, and isn't 'privilege' deeply entwined with your ability?

It's not pretty or P-C, but it's not exactly confusing to me that someone with more opportunities for schooling performs better in an environment labeled 'meritocratic'. It's not "Nature", it's "Nature AND Nurture".

One example that I have seen happening in the UK: take two people with directly comparable excellent university education (same university, same course, same results) - the one who went to the "better" school will be perceived as having more "merit" than someone who achieved the same result from a far less privileged background. To me the latter, having achieved far more, has the more "merit" whereas to others it often the opposite.

I don't get it. If it's the "same university, same course, same results", which one is the "better" school?

I believe that "school" here means "high school" or below.

Correct - I suspect that if you ask anyone in the UK where they went to school they'd tell you where they were before university (i.e. primary or secondary schools).

Is that really the case? The majority of people have very little idea of "better" schools beyond a handful of elite institutions, their local area and those that stick "grammar" into their title. It's also reasonably well established that the better state-funded schools are as good as or better than many private school in terms of academic performance.

I'm seeing "Eton-educated" increasingly used as a pejorative description by inverse snobs too.

I wouldn't say it is widespread and certainly not in the software world - but there are environments where that kind of daft thinking does occur.

Even meritocracy based on IQ would likely be correlated to privilege. IQ is highly heritable.

http://en.wikipedia.org/wiki/Heritability_of_IQ

An article posted on HN had an interesting quote on merit (this time in the context of college admissions):

...the dark little secret about [merit-based] grants. They do not create access to higher education for students who otherwise could not go [for financial reasons]. Merit-based price discounts only help determine which school a student attends. A fully funded need-based aid system is what would create real access for students who would otherwise be shut out or have to trade down to lower quality alternatives.

If correct, this means that at the college level, merit is also highly correlated to privilege.

http://economix.blogs.nytimes.com/2011/02/18/why-does-colleg...

Scores on IQ tests are highly heritable. Whether IQ is highly heritable depends on whether you define it as the result of the test or something that the tests are trying to measure.

A lot of tests tend to embed a ton of middle-upper-class cultural touchstones that you don't even realize are there until you find yourself explaining them to a test-taker who's never been camping or on a boat.

EDIT: The concept of "intelligence" that IQ purports to measure is probably highly heritable as well, of course. I'm just cautioning against drawing too many straight lines from test results, particularly on the topic of diversity and intelligence.

Privilege isn’t just about money – it’s about a collection of gifts individuals receive from society, regardless of merit.

Those gifts include but aren’t restricted to freedom, time and indulgence. And to say that a person can only contribute to an open source project if they have equivalent freedom, time and indulgence from society as our very most privileged members, is to exclude a great number of people – I’d say 99% of people – who are capable and would be wonderful additions to our community.

So to address your question about privilege being entwined with ability: it is dangerous to assume a relationship between the two. I choose to question my assumptions about ability (because I know that I have biases!), because the truth is that I have no idea about other people until they actually do something. And even then, my understanding of someone else's abilities is often limited to only what I directly observe.

Keeping that in mind opens up opportunities between people. And I prefer that mindset to one that assumes a link between privilege with ability.

Could you be specific about what that "collection of gifts" is and how Postgres/OSS needlessly excludes people based on these gifts?

I listed a few: freedom, time and indulgence. More specifically, the freedom to pursue passion in life is often enabled by one's parents having enough money and stability. Time is mostly about leisure time. Having a job that supports work on open source is still not that common, so one is left with figuring out how to do this on your own time - and that depends on you actually having free time because you don't have kids, you only have one job, and you have the background knowledge (socially, tools-wise and background reading) necessary to participate.

So, that's how the whole world works, right? I am not saying that Postgres is unique here. I think we actually do a really great job. We have about 300 contributors every year, and only about 20 committers. That's impressive and shows how the contributors feel about our project. It is prestigious, interesting and fun enough for them to send us their code even though we don't give out commit access, and we subject everyone to a pretty difficult code review.

That just goes to show that if you ask a lot from people, they like it and come back for more. Or at least the people I want to be in a community with like it. :)

I am bringing up to my community that we could do a little better inside of Postgres. We could take a closer look at how we operate, and find ways of inviting people in that don't already fit our mold.

There isn't an easy solution to addressing privilege inside a community, at all. But I was at a conference last week, and a couple people asked me to write my thoughts down about meritocracy. It was not an easy thing for me to do.. I already see a few de-railing comments about gender.

But I continue to get support for thinking and writing about this issue, and ultimately finding more ways of getting more contributors. This is why I continue to contribute, and why I took a moment to make a gentle criticism of a community that I enjoy very much.

I'm confused. From your original post: And to say that a person can only contribute to an open source project if they have equivalent freedom, time and indulgence from society as our very most privileged members, is to exclude a great number of people – I’d say 99% of people – who are capable and would be wonderful additions to our community.

This seems to postulate that OSS excludes 99% of people who are capable of contributing good code. However:

...freedom, time and indulgence...the freedom to pursue passion in life is often enabled by one's parents having enough money and stability. Time is mostly about leisure time. Having a job that supports work on open source is still not that common, so one is left with figuring out how to do this on your own time - and that depends on you actually having free time because you don't have kids, you only have one job, and you have the background knowledge (socially, tools-wise and background reading) necessary to participate.

Now you are discussing people who don't have the free time or knowledge to contribute good code. How would such people be a wonderful addition to a community devoted to writing good code?

You've created a circular argument - if someone doesn't currently have free time or knowledge, how could they ever write good code?

re: free time - I didn't suggest any particular solutions, because the point of this exercise isn't for me to give everyone easy solutions to problems. But I'll offer you one:

More companies could offer time to their employees to work on open source projects. That's something the project itself can help lobby for, and can keep in mind when soliciting help.

That's not a new idea. Many others have said it, and many companies implement it.

I made no argument at all. I merely asked a question.

Excellent points. Also, in many environments, the standards about how different contributions are valued [and hence how ability is assessed] have been defined primarily by those with certain kinds of privilege -- one more reason to question biases about ability!

I prefer that mindset to one that assumes a link between privilege with ability.

I am not assuming it. I was merely pointing out that it could be the case, which I feel is relevant- especially because the author assumed there was no connection out of hand.

In my opinion, before trying to fix a system, you need to confirm it is actually broken. And by 'broken' I don't mean, 'not fair'- I mean, 'not doing what it's supposed to do'.

It's a logical error to look at the outcome of a process and draw conclusions about it with no other supporting data. This error would be innocuous if it wasn't the underlying cause of every inhuman prejudice we've spent the last few centuries trying to rid ourselves of.

If I were to say, "Since there are so few women contributing to open-source projects, it's obvious that women aren't capable of producing much valuable code," the hacker news server would melt from the angry response.

But apparently it's ok to say, "Since there are so few women contributing to Postgres, it's obvious that it's not a meritocracy." Logically, however, these statements are equally erroneous.

You can use this sort of pseudo-statistical handwaving to justify ANYTHING that sounds remotely plausible. That's ignorant, and that sort of ignorance should be opposed no matter the form.

There are exactly zero occurrences of the words "women" or "female" in the article. The only reference to gender at all is a brief and completely inconsequential mention of the fact that your typical Postgres hacker is, among other things, male.

Yet somehow you end up debating the veracity of, "Since there are so few women contributing to Postgres, it's obvious that it's not a meritocracy", a statement neither made verbatim nor in any way implied by the actual article.

Isn't it just a little crazy? Can a writer mention a simple fact without it becoming all about battle of the sexes?

I don't believe that drblast was trying to make this about a battle of the sexes but was just using gender as an easily identifiable example.

The article states that Postgres isn't a meritocracy because the hackers aren't from diverse backgrounds. drblast was attempting to show how this is a logical fallacy - meritocracies aren't ipso facto diverse.

I think it's rather impractical to attempt to guess a stranger's intent. Drblast responded to the article by quoting a hypothetical statement of his own ("if I were to say X") and then quoting a supposed statement from the article ("but apparently it's ok to say Y"). You can interpret that 10 different ways but why? Isn't it simpler to point out that the article never made statement Y in the first place?

Incidentally, I find the point you make - meritocracies aren't ipso facto diverse - worthy of debate. However I don't believe it's the same point as the parent.

The article said that the project was "dominated" by people who are "male," among other things.

The logical error here is the point, not the specific characteristic of whatever group the author is concerned about. The point would be the same whether I had used her "male" example or her "affluent" example. She's making the same error in both cases.

Thanks for your thoughts! Made me think.

I brought up privilege, not gender diversity.

Here is an interesting article by the man who created the term "meritocracy" as part of satirical work:

http://www.guardian.co.uk/politics/2001/jun/29/comment

An interesting quote, which I'm sure I've heard a few times in relation to the recent banking crisis:

"If meritocrats believe, as more and more of them are encouraged to, that their advancement comes from their own merits, they can feel they deserve whatever they can get"

[Edit: As far as I can this concern really applies to organizations like those in the UK public sector where huge salaries are often justified based on rather dubious "merit" - not because of any measure of the value delivered or risks taken]

You'd think anyone with merit would see the flaw in the logic: p implies q, q, therefore p. Maybe the don't have as much as they think.

I think they see the logic as:

p iff q. q. therefore p.

which is a valid inference.

No. They are trained to expect a reward for effort; they get rewards; and then they think it's because of effort.

(Frankly, I think your misreading is a symptom of the same problem.)

Success; that is the theory. Prosperity supposes capacity. Win in the lottery, and you are an able man. The victor is venerated. Have but luck, and you will have the rest; be fortunate, and you will be thought great. Beyond the five or six great exceptions, which are the wonder of their age, contemporary admiration is nothing but shortsightedness. - Victor Hugo

I spend time on the Technical Panel of a security standards body called TigerScheme[1] which is very much a project to which people contribute free time in a similar way to many open source projects (although TigerScheme isn't an open source project, as it doesn't release code). I recently used the phrase 'do-ocracy' to describe it. If you have the time, want to do it, run it past the other people and if there are no objections within a reasonable time frame, do it. Previously Tigerscheme suffered as volunteers wouldn't be available in decision making. I believe this 'do-ocracy' approach has actually enabled Tigerscheme to make decisions quicker and take better ownership of actions.

[1] - http://www.tigerscheme.org/

I'm not very good at many things, but I'm comforted often by the fact that the things I could do, and did, work better (by definition) than the things that the better people could've done, and didn't.

I wrote the entry that started this thread, and an earlier post that mentions do-ocracy in the context of the Portland tech scene[1]. I totally agree with you, and go even further when it comes to the community that's evolved in Portland. I tell people to JFDI. That usually means someone tells someone else about what they're doing. But there's no permission seeking. We have too many people that find fault in the works of others, even before any work has been done.

1 - http://www.chesnok.com/daily/2010/03/17/this-town-needs/

Surely the ability to write great code is closely linked to the ability to earn an income in the top 1% of the world - which according to 'Global Rich List' (lord knows if it's accurate) is around $50,000.

In a field where merit is translatable to wealth, a large proportion of wealthy people in a subgroup points to more meritocracy, not less.

The definition of "meritocracy" that the article provides isn't the dictionary definition of the word; it's the first sentence of the Wikipedia article on the topic where the opening paragraphs are talking about the historical origination of the concept. The dictionary definition reads, "government or the holding of power by people selected on the basis of their ability". The Wikipedia article lists intelligence, credentials, and education as examples of potential measures of merit but the particular measures of merit used in a given meritocracy aren't inherent to the words definition.

Whether or not advancement in a meritocracy is correlated to privilege is a function of what measures of merit are used not some innate property of meritocracies in general. In the context of an open source project one would hope it is a meritocracy of ideas where the merits being judged are something like the quality of the algorithms and code.

Selena's wish to be more open to contributors that have less time to contribute isn't a question of meritocracy it is one of organizational culture; unless the merit on which they are judging people is the amount of raw time they spend on the project vs. the quality of their contributed work.

Yeah, that's a really good point. I am talking about organizational culture. More specifically - structure around participation.

But, at the same time, it seems people often conflate governance and power with organizational culture. Is there a good reference for differentiating the two?

"I don’t believe [meritocracy is] our strongest value, or quality as a community. And, it’s not something that I think embodies what is awesome about Postgres.

Our strongest quality is our ability to create great code."

Let's see... -great code -> merit -transparency -> code improvements (best idea wins)

And frankly, the "privileged" people he's talking about contributing to Postgres make the money and lead the lives they do precisely due to their merits as programmers.

Sounds an awful lot like an organization that promotes the best ideas and gives power based on the merit of the people in it to me.

I'm a she, and you're leaving out all the things that led to becoming a programmer. Which largely had nothing to do with merit.

The point of the article isn't to discount the work involved in becoming excellent, but to point out that the opportunity to do that work at all is a privilege.

I don't particularly care if any one person changes their thinking about this. My post was meant for my community, which is taking the issue of involvement and its relation to privilege seriously.

...you're leaving out all the things that led to becoming a programmer. Which largely had nothing to do with merit.

Is your objection to the idea of "merit" merely reductionism?

I.e., merit (as defined by ability and willingness to code) is comprised of a combination of innate factors (genetic/environmental contributions to intelligence and demeanor) and social privileges (access to education, etc)? This is true. So what?

Ultimately, merit is merely a particular arrangement of atoms. It's hard to see how this reduction is useful.

Most people generally aren't comfortable conflating certain flukes of genetics and social privileges with merit.

Being born affluent, white, and male are the three factors which for the last couple of decades have made you statistically most likely to have owned your own computer prior to college. Because most people don't consider males to have more merit than females, or white people to be more meritorious than minority groups, declaring programming a meritocracy ignores and conceals these important factors in the makeup of the "programmers" social group.

Most people generally aren't comfortable conflating certain flukes of genetics and social privileges with merit.

No one is doing this. I am defining "merit" to be the ability/willingness to write good code.

If the universe has decided to allocate merit (defined as above) to white/asian males, so be it. That doesn't mean that my definition of merit is flawed. It merely means that the universe has not behaved in the manner you wish it did.

My objection is that the use of merit to justify rewards typically seen as an objective measure.

We should carefully identify the reasons for measuring and rewarding a thing - like "we reward excellent code with commit rights", rather than that we "reward people based on merit". The word itself is fraught, as is the philosophy of meritocracy. It's used as a broad brush for many different meanings.

We should just be specific, and say what it is that we are rewarding.

All you are doing is quibbling over definitions. It doesn't matter if you call it a "good-code-writer-ocracy", a"meritocracy" or "wheezlebum" - we are describing the same objective reality.

http://lesswrong.com/lw/np/disputing_definitions/

I'm not quibbling over definitions. And we are not describing the same reality.

The crux of what people mean when they refer to meritocratic systems or organizations has nothing to do with the people in the organization or is related only tangentially to people. What they really mean is that the best ideas win, that the people in the system operate in accordance with a set of values that place the excellence of the end product above the egos of the people, especially the popular or powerful people. Do we have a suitable word to use in place of “meritocracy” and friends?

for those thinking about shaping communities, this is a key takeaway: "What we have to do is create structures that invite people to give what they can, when they can give it." good advice for all of us!

Reply on news.ycombinator.com