It’s a question of its context

I read an article in Harvard Business Review today entitled “I won’t hire people who use poor grammar,” by Kyle Wiens. Wiens assesses job candidates, in part, on the basis of their use of English grammar. He goes so far as to administer a written grammar test to all applicants.

Amusingly enough, the website generated a URL by truncating the title to “i_wont_hire_people_who_use_poo.” I’m pretty sure I wouldn’t, either, unless using poo happened to be part of the job description. “Seeking howler monkeys for stock floor trading positions. Throw your résumé against the wall and see if it sticks.”

Um, okay, where was I? Oh, yeah. Is Wiens’ approach excessive? Ah…wait a second. Should that be, Weins’s? Does it depend on whether you’re in the US or UK? Does it depend on which form your fourth-grade teacher thought was “the rule?” <sigh/> I guess my chances of passing Wiens’ grammar test are low. Oh, wait…is it okay to use faux XML in a narrative? I’m so confused!

Anyway, comments on the article run the gamut from strong approval to strong disapproval. I find myself both agreeing and disagreeing with Wiens.

Let’s start with the points of disagreement. That’s usually more fun.

First, though, I want to commend him for writing an article criticizing other people’s use of grammar. Any time you do that, you know readers will jump all over any minor error you happen to make in the piece. Politics and religion are far safer topics. You would have less poo thrown at you if you made a speech promoting gay marriage to the Republican National Convention. Kudos to him for having the courage to step up and criticize other people’s grammar in public.

People tend to get hung up on the details when they are criticizing grammar. Among the 2000+ comments his article had garnered as of the time I wrote this post, many of them make no substantive statement at all, but merely criticize the grammar and spelling in other comments. Here is one representative sequence of comments:

  1. There are an astounding number of comments posted in support of the author’s viewpoint that contain grammatical errors.
  2. I think you mean ‘there is an outstanding number of comments’. Number is sungular. Sounds wrong, but it’s right.
  3. I believe the spelling is “singular”.

Yeah, thanks. That was worthwhile. Speaking of attention to details (well, I haven’t spoken of that yet, but I’m going to later (yes, for you sticklers out there, “to later” is obviously the infinitive of the verb, “later”)), the first guy wrote “astounding,” not “outstanding.” And, by the way, the singular doesn’t sound wrong in that context. And that probably wasn’t a spelling error, but a typo. The U is right next to the I on a qwerty keyboard. So…yeah, right, whatever.

Now, on to the disagreement. Wiens refers to himself as a grammar stickler, a term he attributes to Lynn Truss. I refer to such people as pop grammarians. A grammarian is a linguist who studies grammar and syntax. Grammarians consider formal grammars to be descriptions of the way people actually use living languages. The grammar of a language changes as the language evolves. A pop grammarian is not a linguist. It is a person who considers formal grammars to be prescriptions — rules that must be followed when using a language.

Rules are not evil. If there were no rules at all, then we wouldn’t be able to understand each other even as poorly as we do. Yet, there is no universal agreement about certain grammatical details. When is a comma necessary? Sometimes it’s optional; sometimes, it’s also optional. I can think of one, two, and three reasons to use a comma; and I can think of one, two and three reasons not to use one. When taking Wiens’ grammar test, it’s a coin-toss whether I answer a question about commas in the way he happens to believe is a “rule.” He might as well save the time and trouble and just toss a coin.

The people who work for Wiens write for a living. It’s very important that they understand basic, standardized “rules” and conventions for English. Wiens points out that an understanding of grammar is important for other businesses, too. He has a point.

But he overreaches when he writes:

In the same vein, programmers who pay attention to how they construct written language also tend to pay a lot more attention to how they code. You see, at its core, code is prose. Great programmers are more than just code monkeys; according to Stanford programming legend Donald Knuth they are “essayists who work with traditional aesthetic and literary forms.” The point: programming should be easily understood by real human beings — not just computers.

Well, you see, at its core, code is not prose. Code is code. This rambling nonsense you’re looking at right now is prose. My prose rambles; my code doesn’t. Code has to be understood by other programmers, not by real human beings. (No offense, guys; you know what I mean.) That means clean code (not prose), in the sense described in books like Clean Code and Code Complete. With all due respect to Donald Knuth, programmers are not “essayists” and programs are not “literary forms.”

Wiens’ grammar test might be appropriate for someone applying for a job as a writer. It’s not appropriate for someone applying for a job as a programmer. If he’s using the test to screen applicants for programming jobs, then I have to wonder whether he understands enough about programming to know how to assess job candidates.

That said, I do agree with Wiens to an extent. There is value in following standard conventions in written language in a business context. One of Wiens’ pet peeves is the recurring question of “its” and “it’s.” I completely agree with him about this one. It’s astonishing that millions of native speakers of English go through many decades of life without understanding this simple point. The possessive form of “it” has no apostrophe. “It’s” is a contraction of “it is” or “it has.” One could write, “it’s not rocket science.” One cannot write, “its not rocket science.” (Well, one could. I just did. But you know what I mean. And, yeah, you’re not supposed to begin a sentence with “but” or “and.” We could nitpick all day.)

It’s a question of its context. Since the purpose of written English is to communicate, it’s rather important that we don’t say something completely different than we intended. “It’s” is completely different from “its.” Consider this example:

Stop its momentum!

That’s clear enough. What if it were written this way instead, either inadvertently or because the writer misunderstood the “rule”:

Stop it’s momentum!

What does that mean? It could mean:

Stop it is momentum!


Stop it has momentum!

“Is” might make sense in an obscure context that isn’t evident from the sample sentence alone. “Has” seems to come a bit closer to making sense, although it’s not very fluid English. Could they have meant “it’s got momentum?” Maybe you’re trying to warn someone that an object that has momentum is approaching them, and they had better stop walking so that the object won’t strike them. (Yes, that’s right: I used the plural them in a singular sense to avoid the clumsy construction, he/she. So, shoot me.) This would be a good place for a comma:

Stop, it has momentum!

…or, to convey a greater sense of urgency:

Stop! It has momentum!

…or, for a more dramatic presentation:

For God’s sake, stop! Can’t you see it has momentum? Oh, the humanity!

But all they really meant to say was

Stop its momentum!

I agree with Wiens. I don’t think it’s asking too much of people to use “its” properly.

I disagree with Wiens. I don’t think it’s reasonable to be a “stickler” about pleonastic niceties. His purpose in testing job applicants on grammar seems to be to ascertain whether the applicant pays attention to details. (See? I told you I was going to mention that.) I doubt there is a correlation between paying attention to grammatical rules and paying attention to rules of, say, programming or accounting. It’s (“it is”) a false measure.

There is, however, a point of diminishing returns in accepting sloppy grammar. In my opinion (optional comma) we pass that point when the poor grammar impedes clear communication. Communication problems often generate unnecessary extra work. That is a cost no business needs. Let me share a couple of anecdotal experiences.

In the 1980s, I was working on a large-scale back-end credit authorization application. A group of about 70 programmers maintained and enhanced the application. One of the senior people in the group was a woman from Vietnam. English was her fourth language. She was very smart (she spoke four languages, after all), but her English grammar was, shall we say, pretty funky. She added a comment to the program that was first in a series of modules that processed incoming credit authorization requests. The comment read (as well as I can recall):

Required for the sub-sequence module.

It became sort of a game for us to watch the reactions of new hires when they encountered this comment. Invariably, they would struggle for a couple of hours and then ask for help. “Okay, I give up. What the hell is the sub-sequence module? I can’t find it anywhere!” Then we would all have a good laugh. What she intended to write was this:

Required for subsequent modules.

Most amusing, indeed…until you think about it from the company’s point of view. That place was a sweat shop. Of the 70 programmers, about 15 of us had the will to stay on for an extended period (I was there for three years). The other 55 positions had high turnover. So, that single badly-written comment cost the company approximately (3 new contractors per week * 2 hours confusion time * 4 people standing around snickering * $65/hr * 52 weeks) = $81,120/year. Not a hell of a lot in the grand scheme of things, but enough to fund a small-scale project. How many other grammar-related communication issues occurred in that shop in the course of a year? What would the total cost of confusion be? It’s anyone’s guess. (“It’s” means “it is,” in that case, in case you were wondering.)

Much more recently (within the past couple of years), I was helping a client incorporate the practice of specification by example into their process, and introducing Cucumber, a tool to support that approach. Before long I noticed that the Gherkin scripts contained numerous redundant clauses; different wording expressing the same intent. The clauses varied in their use of definite and indefinite articles, in their use of singular and plural nouns, and in a few cases in their word order. There was also a bit of carelessness regarding consistent domain language, but that’s a different issue. The differences were the result of the particular grammatical abuses perpetrated by contractors from different countries where English serves as a common language for education and business.

There were between 5 and 8 variations on the same statement. That means that if the group eventually developed 6,000 examples with an average of 2 given clauses, one when clause, and 2 then clauses each, they would have (6,000 examples * 5 Gherkin statements * 6.5 grammatical variants on average) = 195,000 unique steps to maintain, including an unmanageable number of redundant ones. Had they followed standard conventions of English usage and grammar, the same number of examples would comprise a maximum 30,000 statements, none of which would be redundant. In reality, there would have been fewer than 30,000 unique steps, as the better-crafted statements would lend themselves to reuse. I would guess the total number of steps could be as low as 300. Far more manageable, far less error-prone, far less confusing.

The way languages evolve in general use and the way languages are used in a business context are two different issues. Many populations around the world use English for one reason or another. Often, they use it as a common language when their country has many different native tongues; for example Nigeria and India. Sometimes, people use it as a common language for business, diplomacy, and education because it is more widely-spoken than their national language; for example, Estonia and the Netherlands. In some countries, technical professionals learn passive English (reading, understanding) so they can consume technical material, but aren’t too interested in active knowledge (speaking, writing); for example, China. Whatever the reason for using English, once people start using it they change it. This means English is a living language that is still evolving.

The ways people are adapting English in different countries enrich the language. The variants spoken in Scotland, Jamaica, Australia and elsewhere are all quite different and all introduce unique elements to the language. They make English more diverse, more colorful, more vibrant. But business use is not general use. For business use, it makes good sense to follow a few basic rules and keep things standardized. Otherwise, we end up with 195,000 steps for Gherkin scripts and the never-ending hunt for the elusive sub-sequence module.

A real grammarian would find the evolution of a language interesting, but neither right nor wrong. A pop grammarian can only see the violations of grammatical “rules.” A real grammarian notes that native speakers started to treat any and none as plural in the late 20th century, and is delighted to witness the real-time evolution of a living language. A pop grammarian simply marks the answer “wrong” and denies the job applicant a fair opportunity to earn a living.

While I do think that we need to follow standard conventions in a business context, my reason for thinking so has to do with avoiding the needless cost of the extra work that is caused when written statements are unclear to the reader. I don’t think it has anything to do with treating code as prose.

One thought on “It’s a question of its context

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s