In Canada, the law against drunk driving says that you are not allowed to be in "care or control" of a vehicle while intoxicated. This allows the police to enforce the law against people even if they are not actually driving. That is, it allows them to intervene before the driver actually becomes a danger. And, interestingly, though "care or control" does have something to do with the intention to drive, it can be attributed to a driver even if he or she insists she was in the car for some other purpose.
Why is this important for writers? Well, if my hunch from last week is right, i.e., some writers are to their writing as some drinkers are to their drinking, namely, "out of control", then there is also a sense in which this is an issue of broader public concern. The writer who is unable to control the writing process is a "danger" to other writers. Drunks, after all, are notoriously "careless" about their surroundings, whether social or material. They hurt other people's feelings, and therefore not only their own social relationships, but the ability of others to form them. And they sometimes, of course, hurt other people physically. That's where the problem of drunk driving comes in.
Is undisciplined writing really just as bad as drunk driving? Of course not. Let's keep in mind that undisciplined drinking is not in and of itself a problem; the problem arises when undisciplined drinking is combined with serious activities (like driving) that can have consequences for other people. Writers who produce texts for publication are putting themselves "in care and control" of a text that other writers will have to deal with. If they are not actually very careful, or very much in control of their writing process, they are likely to make mistakes that will affect their peers. The most obvious example is plagiarism, arguably the most serious form of carelessness in writing, but we can also talk about everything from misreading to misspelling. All mistakes "count" in academic writing because they undermine the reliability and usefulness of a text.
This is why it is so important to give yourself conditions (in time and space) under which to write carefully. It is today illegal in many countries to talk on the phone while driving and the reasons are similar to the laws against drunk driving. It is about being a careful, controlled participant in traffic, which is a matter of public concern. The same goes for writing. While you are working on a text for publication, you should be thinking about that text very seriously, and not a lot of other things. Don't talk on the phone while writing a paper, needless to say. And you should not be under the influence of pressures (like promotion and tenure) in the moment (put that out of your mind while writing). You also shouldn't be drunk, which, unfortunately, does not go without saying. The best way to ensure these conditions is to control the space-time coordinates of your writing process. Know when and where you will be committing words to the page for publication.
That qualification ("for publication") is important. The writing that you want to have "care and control" of in this serious sense is your scholarly writing, i.e., the writing you are doing for others. Here you are building a particular kind of relationship, both social and material, with your surroundings, and you want to do this reliably, deliberately, carefully. There is a lot of writing that you can do much less carefully without causing too much trouble, mainly because it is read in the same spirit. Consider the difference between emails, blog posts, journal entries, and "thought writing", on the one hand, and the work you do on a book or a journal article, on the other.
The reason drunk drivers don't always cause damage is that they often have a very "routine" drive home, offering nothing they need to react to quickly. There are no critical moments, so they don't need their critical faculties to be functioning at the top of their game. The same is true of a text you are writing for publication. The weaknesses of a sloppily written text may not be apparent on a first reading or even during the review process. But once other scholars try to engage with it in a critical, detailed manner, issues will more than likely turn up. It may not cause any harm, of course, mainly because the writers around you may just start keeping their distance. Indeed, you are making an extra demand on their critical faculties.
Keep in mind that when you are writing with the intention to publish you are "in care and control" of something that other people have an interest in. Take care. Be in control. Or you will make others wary.