Category Archives: Punctuation

Posts that discuss punctuation

The Oxford Comma, or the Limits of Expectation

Why talk about the Oxford comma? Surely everything has already been saidsung, or drawn. But why even have a blog if you can’t use it to share your great love of the Oxford comma? More importantly, its use comes up in all my classes; if I say ‘comma’ to a room of graduate students, someone will immediately ask whether it’s right or wrong to put a comma after the penultimate item in a list. I can explain both sides of the issue, but I cannot, in the end, answer the ‘right or wrong’ part of the question. Which leads me to another reason to talk about it: the very fact that there’s no simple answer to the question is instructive for the way we think about issues of style in our academic writing. Before delving into this topic, let’s look at what the Oxford comma actually is.

The first thing to note is that it is properly called a ‘serial comma’, the comma that appears before the conjunction in a list.

Peace, love, and happiness

Do you put a comma after ‘love’ or don’t you? That’s what my students want to know. Actually, that question is pretty easy—I do, always. The harder question—and the one they care about—is whether you should put a comma there. That’s the question that can’t be answered on the grounds of correctness. Both are correct, unless you are writing for a particular journal or press with a stated preference. Since most of my students are primarily concerned with writing their theses, they often have to make this decision themselves. And so they want a definitive answer. After I give a full accounting of the subtle pleasures and perils of the serial comma, someone is sure to say, while stifling a yawn, ‘so should we use it or not?’. Again, either way is fine, but here is a summary of my non-answer.

1. Despite the name and the song, the Oxford comma isn’t more formal. Students often tell me that they think it’s inessential but probably a good idea for a formal piece of writing. In truth, however, the serial comma is neither formal nor particularly stuffy. It’s not even more British: its presence in the Oxford style guide notwithstanding, it’s much more common in the US than in the UK. The only way in which it’s more formal is that it isn’t generally used in newspapers where narrow columns put space at a premium.

2. All the evidence about ambiguity can cut both ways, but in ordinary academic writing—in which internally complex list items are routine—the serial comma will help more often than it hinders.

This theory of community engagement addresses tensions within the spheres of politics, arts and culture and finance.

This theory of community engagement addresses tensions within the spheres of politics, arts and culture, and finance.

In this case, I think that using the serial comma is the easiest way to show what goes with what. I also recommend using semicolons when list items get unruly, and those are always used serially. Overall, avoiding ambiguity is our responsibility as writers. If commas can’t help us, then we need to reword; that is, there are many instances in which the problem is not the presence or absence of the serial comma, but rather the awkwardness of the list itself.

During diagnosis, treatment and monitoring of a patient’s pathologies, measurements of medication levels are often essential.

During diagnosis, treatment, and monitoring of a patient’s pathologies, measurements of medication levels are often essential.

Measurements of medication levels are often essential during diagnosis, treatment, and monitoring of a patient’s pathologies.

While the second version avoids the obvious potential for misreading found in the first, the third may be better overall. I’ve written it with the serial comma, but it would, of course, be fine without.

3. Even if neither option is wrong, it’s still a good idea to make a decision. That is, just because both options are correct doesn’t make flipping back and forth desirable. Consistency is a useful principle when making style decisions. Doing the same things the same way every time is a kindness to your readers—since they will be saved the bother of wondering about stylistic inconsistencies–and a kindness to yourself—since you’ll be saved the bother of wondering what’s right each time. Some people will use the serial comma only when ambiguity might result from not using it, but I think that practice can potentially confuse the reader.

It took me a long time to get over my belief that the serial comma was inherently better. I believed this for years, and that was before I went to work for Oxford University Press, where it is house style. My editorial eyes are deeply attuned to it, and its absence trips me up every single time. But that just shows the limits of our own expectations. The serial comma is only better in a world where it is expected to appear. The only surefire solution is a world in which everyone does everything the same way, which is implausible and slightly creepy. The best we can do is choose our way and stick to it consistently, so our readers—consciously or unconsciously—become accustomed to our punctuation habits. Again, this is why I don’t like the practice of using a serial comma only when it’s obviously beneficial; inconsistency undermines our reader’s ability to becoming habituated to our writing style.

While the Oxford comma may seem more like a punchline than a punctuation mark at this point, I think that there is something important in this conversation. Writing requires us to make choices far more often than it demands simple rule following. And those choices should be made based on our best understanding of style conventions and reader expectations. The serial comma shows us the limits of expectation, but it also confirms the importance of making informed and consistent decisions about academic writing style.

Full Stop

As I was proofreading something recently, it occurred to me that discussions of punctuation and grammar contain relatively little about the period. If you go looking for information on using periods, you will mostly find technical advice about formatting. Such editorial advice is crucial, but not necessarily of primary interest to the writer. The obvious reason that writers don’t worry much about periods is that their use is pretty straightforward: we use a period after a complete sentence. The period has none of the apparent complexity of the semicolon, the colon, the dash (more on the dash here), or the comma (more on the comma here, here, here, and here). But that relative simplicity doesn’t mean that the decision to use a period isn’t consequential. A period makes evident key authorial decisions: whether to say a lot or a little before stopping; whether to be assertive or interrogative when raising questions; or whether to insert a full stop or just the rolling stop achieved by other types of punctuation. I certainly don’t want to problematize the one piece of punctuation that everyone feels good about, but the ways that we stop in our writing are worth some consideration.

Short vs. long sentences. Any time we use a period, we are deciding whether we want a short or long sentence. If we are generous with our periods, we will be writing clipped sentences; if we are stingy, on the other hand, we will end up with sprawling sentences. Neither is better. The previous sentence, for instance, is short because I want to emphasize it. And its brevity is enhanced by the fact that it follows a longer set of sentences with a semicolon framing their double-barrelled structure. It generally works well to use short sentences for emphasis and longer ones for depth. If we go too far in either direction, the reader will begin to suffer. Too many long sentences coming one after another—even if they are individually well structured—can be fatiguing for the reader. Too many short sentences will disorient the reader since not everything can warrant special emphasis. Those short sentences will inevitably contain unimportant things; the reader’s confidence in you can be undermined if you appear to be emphasizing the insignificant. Creating a healthy mix of short and long sentences can be the best thing for your paragraphs.

Direct vs. indirect questions. What is the impact of asking a question as compared with stating that a question exists? When raising questions in our work, we have to decide whether to use a direct question—i.e., one that ends with a question mark—or an indirect question, i.e., one that ends with a period. When we use too many direct questions, our readers may start to see us as having more questions than actual strategies for answering those questions. An indirect question gives us the ability to articulate a question, which is obviously essential to academic writing, while also indicating our plans for addressing the question.

Full stop vs. rolling stop. As a hopeless over-user of colons and semicolons in my own writing, I have a particular interest in this question. During editing, I often find long strings of sentences, all of which are connected by semicolons. An over-reliance on the rolling stop is a bit like an over-reliance on transition words: it tries to impose flow rather than establishing it. Genuine flow will involve the use of some colons and semicolons; there’s nothing wrong with nudging your reader towards the next sentence. The semicolon says that the next sentence is intimately connected to this one; the colon says that the next sentence completes what is being said in this one. Those relationships are emphasized by the punctuation choice, but the punctuation alone won’t create that relationship. The choice between a full and a rolling stop should be consciously based on an understanding of the nature of the connection between sentences.

In sum, if you understand what is requisite for a complete sentence, your use of periods will be perfectly correct. But thinking about how and why we end sentences can still be a useful avenue for improving the flow in your writing.

Commas and Relative Clauses

Our task for today is to understand how we punctuate relative clauses. In the simplest terms, a relative clause is a clause that begins with a relative pronoun (which,  that, who, whom, whose). Let’s begin by looking at this example of a sentence with a relative clause:

CNCP patients, whose complaints of pain are not adequately addressed, start to display aberrant drug-related behaviours that are mistaken for addiction.

This sentence—taken directly from student writing—is not incorrect as written, but it doesn’t say what the author intended. Here is what the author meant to say:

CNCP patients whose complaints of pain are not adequately addressed start to display aberrant drug-related behaviours that are mistaken for addiction.

The difference? The second version of the sentence shows that it is about a subgroup of CNCP patients ‘whose complaints of pain are not adequately addressed’. There are many CNCP patients in the world and only some of them suffer in this manner. The first version says that all CNCP patients have complaints of pain that are not adequately addressed. Because of the commas, we have to read the relative clause as supplementary information about all CNCP patients. Technically, our first sentence could be reworded as follows:

CNCP patients [all of them] have complaints of pain that are not adequately addressed. CNCP patients start to display aberrant drug-related behaviours that are mistaken for addiction.

Rewording the sentence in this way reflects the fact that the original sentence portrayed the relative clause as supplementary. But the author’s intention was not to provide extra information about this group of patients; instead, the author wanted to define a particular group of patients under discussion. The lack of commas in our revised version indicate that the information following the relative pronoun is integral to the antecedent noun:

CNCP patients whose complaints of pain are not adequately addressed start to display aberrant drug-related behaviours that are mistaken for addiction.

The bolding emphasizes the integration of the relative clause. This integration is conveyed to the reader by the absence of commas. When we do use commas, we are telling the reader that we are providing supplementary information.

I chose this example because it is easy to see—even without being familiar with the subject matter—that the punctuation in the original sentence was probably misleading. One of the great difficulties in explaining how to punctuate relative clauses is that context matters. I always tell students to take whatever I have said about punctuating relative clauses home with them: only in applying those principles to their own sentences—sentences that they themselves fully grasp—will they come to understand whether a relative clause is integral or supplementary.

If you are familiar with this topic, you will notice that I am not using the traditional terminology (restrictive and nonrestrictive) or the usual variants (defining and non-defining, essential and inessential, identifying and non-identifying). It is possible, of course, to explain what is meant by these terms, but I have never found the common terminology to be particularly intuitive. More recently, I have noticed people using the terms integral relative clause and supplementary relative clause. I find these terms to be more intuitive, which is why I have started to use them in my classroom teaching. I would be interested to know if anyone has thoughts about whether this different terminology is helpful or just confusing.

Now let’s look at some more examples to reinforce the distinction between integral and supplementary relative clauses.

There are many narratives that can be used to illuminate the psychological concept of extraversion.

The relative clause is integral to the meaning of ‘narratives’. The sentence isn’t just telling us that there are many narratives. It is telling us that there are many narratives that can be used in a particular fashion.

The philosophical approach that is articulated by Rorty will set the tone for the proceedings at the conference.

This sentence is telling us what will set the tone for this conference. And it isn’t just any philosophical approach: it is the philosophical approach that is articulated by Rorty. Again, the relative clause is integral to the meaning of ‘philosophical approach’. Now let’s look at some examples of supplementary relative clauses:

Given the educational conditions in Malawi, which is located in eastern Africa, creative teacher training programs are essential.

Using transactional memory, which requires special hardware or software support, will address the problems associated with using locks.

Theorists argue that gender equity, which is defined here in economic terms, is a crucial component in any attempt to address the global AIDS crisis.

In each of these cases, the antecedent of the relative clauses is completely sufficient without the relative clause. A country is a useful example since it is easy to see that you don’t need any additional information to know what is meant by Malawi. Its location within its continent is obviously supplementary information. Likewise, ‘transactional memory’ is a fully defined term: the fact that it requires special hardware or software support is extra information. Take that information away and the term itself is just as informative. In the third example, even though the supplementary relative clause claims to be defining ‘gender equity’, it is doing so in a supplementary way. The sentence is telling us that gender equity is crucial and it is also clarifying what gender equity means in this context.

Here is a final example, one that gives three different versions of the same sentence:

The articles, which stem from the 1970s and the early 1980s, show Lefort intent on persuading the reading public about the totalitarian nature of the Soviet Union and the countries of the Eastern bloc.

The articles that stem from the 1970s and the early 1980s show Lefort intent on persuading the reading public about the totalitarian nature of the Soviet Union and the countries of the Eastern bloc.

The articles which stem from the 1970s and the early 1980s show Lefort intent on persuading the reading public about the totalitarian nature of the Soviet Union and the countries of the Eastern bloc.

The first two sentences follow the pattern I have been discussing. I chose this example because it shows how easily ambiguity can arise when we’re not clear about the punctuation we need. The first sentence is discussing a group of articles and using its relative clause to give us extra information about when they were written. The second sentence, on the other hand, is using its relative clause to identify a particular subset of articles. The implication of the first sentence is that all the articles were written in the 70s and early 80s. The implication of the second sentence is that there is a broader group of articles (presumably spanning a broader time frame); the author is drawing your attention to a subset of that broader group. Needless to say, it is important for the author to clarify which is meant. In my own experience, the decision about how to punctuate relative clauses often helps me to clarify my own meaning. Similarly, in discussing this issue with students, it often emerges that they aren’t quite sure what they were hoping to convey through their punctuation choices.

But what of the third sentence? Is it the same as the second sentence or is it different? In other words, is it okay to use ‘which’ to introduce an integral relative clause? Yes, it is. But while I would love to leave it at that, I feel I should say something about how I view this issue. The good news is that we have already covered the important part: you must signal your intention to your reader through your use of commas. If the information is integral, skip the commas; if, on the other hand, the information is supplementary, show that with your use of commas. Simple enough. But you do need to choose a relative pronoun and, for many, that decision raises a certain anxiety. When I ask students about their habits in this regard, I get a range of replies (often involving something a high school English teacher once said): guessing and then feeling bad; turn taking (first ‘which’, then ‘that’); using ‘which’ because it is more formal; never thinking about it. For a fairly typical prescriptive discussion, see this post from APA Style. For a more nuanced, historical view, try Stan Carey’s excellent post on this topic (as usual, Carey also provides a very helpful roundup of what others have said about this issue).

Given the general uncertainty this topic engenders, what should we do? My own preference—and that is all it is, a preference—is to use ‘that’ without commas and ‘which’ with commas. The first part of this practice is unexceptionable: nobody uses ‘that’ to introduce supplementary information. It is the second part that causes heartache. Look at this simple table:

integral supplementary
that YES NO
which ?? YES

My preference is to replace those two question marks with a ‘NO’. Not, to repeat, because I think this use of ‘which’ is wrong, but only because I like the clarity and simplicity of reserving ‘which’ and ‘that’ for different uses. I start with the important question—do I need commas or not?—and then use that as the basis for my decision about what relative pronoun to use. I explain this to students in just these terms: once they have sorted out the important issue of how to punctuate, they are free to choose their relative pronouns however they wish. But I do stress that this distinction is often treated in more absolute terms in advice on scientific writing. Whether or not this is true across the board, I do suggest that students preparing scientific papers consider reserving ‘which’ for instances in which they are using commas to convey supplementarity. For me it all comes down to this principle: if our audience might find a particular usage to be ambiguous—even if we know that it is perfectly acceptable—it can make sense to avoid that usage.

There is much more that could be said, but this post is already far longer than a blog post should be! If there is anything that you see as needing further explanation or elaboration, I would love to hear about it in the comments.

This post is the fourth in a series of posts on comma use. The first post dealt with commas and coordinating conjunctions. The second dealt with non-standard commas and punctuating for length. The third dealt with the importance of knowing when you need a pair of commas.

Pairs of Commas

This post is the third in a series of posts on comma use. The first post dealt with commas and coordinating conjunctions. The second dealt with non-standard commas and punctuating for length. Today’s entry will discuss the way that some commas work best in pairs. Compare these three sentences:

This innovative new technique, developed by Woljert, has altered the way this surgery is performed.

This innovative new technique, developed by Woljert has altered the way this surgery is performed.

This innovative new technique developed by Woljert, has altered the way this surgery is performed.

The first sentence clearly conveys its meaning to the reader: there is an innovative new technique that has altered the way some surgery is performed. The reader is also given supplementary information: this technique was developed by some researcher named Woljert. This element (‘developed by Woljert’) either takes two commas (as in the first sentence) or takes no commas (‘This innovative new technique developed by Woljert has altered the way this surgery is performed.’). A single comma here (as in both the second and third sentences) will throw off your readers because it doesn’t clarify the grammatical role of the adjacent information.

When clauses like this appear at the beginning or end of a sentence, the need for paired commas is obviated:

According to Chen, this new technique is very valuable.

This new technique is very valuable, according to Chen.

This new technique, according to Chen, is very valuable.

In each of the first two sentences, an additional comma becomes unnecessary because of the placement of the clause at the beginning or end of the sentence. In the third sentence, we see the need for two commas to clarify the role of this brief interruption. (Of course, you will have noticed that the first sentence sounds much better than the other two. You don’t generally want to give this sort of unimportant information such a prominent place at the end of the sentence. Similarly, you don’t generally want to interrupt a sentence in this fashion unless the interruption itself is significant.)

This use of one comma instead of two (or none) isn’t a particularly grammatically complex issue, but it is a frequent occurrence in the student writing I see. And while it isn’t fatal, it does make your reader’s life more difficult. In lots of cases, this mistake may just be random carelessness; it’s certainly easy to miss a comma here or there. But if you repeatedly use a single comma when two (or none) are needed, you  may be experiencing some confusion between restrictive and nonrestrictive clauses: not knowing whether to use no commas (as you would for a restrictive clause) or two commas (as you would for a nonrestrictive clause), maybe you split the difference and use just one. Come back for the next comma post, in which I will try to sort out the difference between restrictive and nonrestrictive clauses. As I will stress then, this distinction is most crucially an issue of punctuation, but I will also touch on the persistent dilemma experienced by writers trying to choose between ‘which’ and ‘that’.

Commas: Punctuating for Length

This post is the second in a series of posts on comma use. The first post dealt with commas and coordinating conjunctions.

Today’s topic is the practice of putting commas in sentences as a response to the length of the sentence. Consider this example:

The purpose of these focus groups was to improve the understanding of the nature of risk and autonomy during outpatient treatment, and focus on exploring the role of the hospital and the professional team in identifying and balancing treatment efficacy and patient comfort.

The comma after ‘treatment’ is, as your grammar book will tell you, unnecessary. Indeed, putting a comma there might cause some readers to think that ‘focus on exploring’ is the start of an independent clause. To illustrate the redundancy of this comma, consider the same punctuation pattern in a very simple sentence:

I went to the store for milk, and eggs.

This sentence will look wrong to most of us since it separates ‘milk’ and ‘eggs’ when they are obviously meant to go together. There is no reason for a comma between ‘I went to the store for’ and ‘eggs’. We wouldn’t say ‘I went to the store for, eggs’. (We would, however, say, ‘I went to the store for milk, eggs, and avocados’ because we punctuate lists of three or more items differently than we do pairs. For more on lists, see here, here, and here.)

What, then, do we think about the comma usage in our first example? The grammar books suggest it is wrong. The analogy with a short sentence suggests that it is wrong. (Of course, I would certainly acknowledge that it is possible to refuse the logic that says ‘wrong in a short sentence, therefore wrong in a long sentence’. In fact, we do sometimes use punctuation differently in short and long sentences; Veni, vidi, vici invariably comes up in such discussions.) But whether it is wrong or not, this is a pattern of comma use that you will see in lots of respectable places.

My response to this uncertainty is to characterize this usage as non-standard rather than as wrong. This assessment alerts students to the fact that the usage may be ambiguous for some readers and may be out of place in some sentences. However, to me, the most salient thing about this type of comma use is that it is generally better to leave it in place than to remove it without further alteration of the sentence. Simply removing it may make your sentence more ‘correct’ but will likely also make it harder to read. Let’s look at our first example again, now with the non-standard comma removed:

The purpose of these focus groups was to improve the understanding of the nature of risk and autonomy during outpatient treatment and focus on exploring the role of the hospital and the professional team in identifying and balancing treatment efficacy and patient comfort.

You can immediately see that this isn’t a good solution. The new sentence is much harder to read and, in fact, some might even read ‘outpatient treatment and focus’ as a single phrase and thus become confused about the sentence structure. What are our options when a pattern of comma usage isn’t standard and yet removing the comma only makes things worse?

The simplest revision would look something like this:

The purpose of these focus groups was both to improve the understanding of the nature of risk and autonomy during outpatient treatment and to focus on exploring the role of the hospital and the professional team in identifying and balancing treatment efficacy and patient comfort.

Using ‘both’ serves to alert the reader that a pair is coming; since the reader is expecting the writer to offer a dual purpose, the reader will be anticipating a compound structure. The repetition of the ‘to’ (‘both to improve and to focus’) will be another hint to the reader as to how to read the sentence. Notice all the ‘ands’ in our example sentence; repeating an element like ‘to’ signals to the reader which ‘and’ is doing the heavy lifting.

Another option would be to restructure the sentence so that it explicitly anticipates the two purposes:

These focus groups had two main purposes: to improve the understanding of the nature of risk and autonomy during outpatient treatment and to focus on exploring the role of the hospital and the professional team in identifying and balancing treatment efficacy and patient comfort.

In this case, the reader encounters ‘to improve’ and then ‘to focus’ with a prior understanding that there are two purposes; by providing that explicit information, the writer makes very certain that the reader will know how to read the sentence. The only downside to this structure is the possibility that it could overemphasize something that might actually be trivial:

The trip to the store had two main purposes: milk and eggs.

But in our original example, the writer seems to be giving important information about the current research, suggesting that the extra emphasis provided by the colon is warranted.

In sum, I think it is valuable to be aware of this pattern of comma use as non-standard. With that knowledge, you may go on using the comma in this way or you may choose to rephrase. My point is that, in many cases, the optimal solution won’t be to remove this comma without also altering the sentence in other ways. In some cases, of course, simple removal will work, but in more cases, when we punctuate for length, we are responding to a valuable intuition that the sentence won’t work without some additional aid for the reader.

Our next comma post will look at the importance of understanding when a comma can stand alone and when it needs a partner.

Commas and Coordinating Conjunctions

I am going to start this discussion of commas by showing a simple pattern of comma use:

Some educators believe in using inductive methods in the classroom, and others maintain that a ‘top-down’ approach is more effective.

In this compound sentence, the comma separates the two independent clauses, indicating that we must read each one separately; the ‘and’ following a comma tells us that the next word (in this case, ‘others’) is the beginning of a new independent clause. The crucial issue here is the presence of a coordinating conjunction. A coordinating conjunction is strong enough to join two independent clauses with only a comma. There are seven coordinating conjunctions: and, but, so, or, nor, for, yet. Only those seven words can give us the compound-sentence-with-comma pattern found in the above example. I have listed them in a way that approximately reflects the frequency of their appearance in this capacity. Obviously, these are words with many other roles to play in our writing, but I am speaking here of their use as coordinating conjunctions. We use ‘and’ and ‘but’ all the time; we use ‘so’ and ‘or’ often; and we rarely (outside of literary writing) use ‘nor’, ‘for’, and ‘yet’. We can also write them in a different order so as to get a mnemonic: for, and, nor, but, or, yet, so. The resulting FANBOYS, despite being a somewhat silly word—I confess to having spent a fair amount of time rearranging those seven letters in an attempt to find a more dignified term—is a handy way to check if a sentence can be punctuated as a compound sentence.

Now that we understand this simple pattern of comma use in compound sentences, we can look at some related comma errors.

One such error is the comma splice: the placement of two independent clauses together with a comma and NO conjunction whatsoever. Here is an example:

The simulation of physical systems is a crucial part of scientific discovery, experience shows that conducting this simulation precisely and efficiently is essential.

This type of error—which is relatively rare—can be easily fixed in a number of ways, including the simple addition of ‘and’ before ‘experience’:

The simulation of physical systems is a crucial part of scientific discovery, and experience shows that conducting this simulation precisely and efficiently is essential.

What if we had attempted to correct this sentence by replacing the comma with an ‘and’? Note the ambiguity of a compound sentence without the comma:

The simulation of physical systems is a crucial part of scientific discovery and experience shows that conducting this simulation precisely and efficiently is essential.

Here it would be easy to read ‘scientific discovery and experience’ as a single phrase and thus miss the true structure of the sentence. Remembering that we always need a comma before a coordinating conjunction in a compound sentence will save us from this potential ambiguity.

Lastly, I would like to look at a related comma error that is very common: the practice of using commas to separate independent sentences joined by conjunctive adverbs (e.g., however, therefore, accordingly, finally, instead, nevertheless, specifically, thus) or transitional expressions (e.g., equally important, for example, in fact, on the contrary, on the other hand). Compare these two sentences:

Low levels of ROS are used in redox signalling reactions that are essential for cellular homeostasis, but high levels of ROS initiate an intracellular response that leads to oxidative stress.

Low levels of ROS are used in redox signalling reactions that are essential for cellular homeostasis, however high levels of ROS initiate an intracellular response that leads to oxidative stress.

The first sentence is correct, but the second one is not. Here is a corrected version of the second sentence:

Low levels of ROS are used in redox signalling reactions that are essential for cellular homeostasis; however, high levels of ROS initiate an intracellular response that leads to oxidative stress.*

In order to understand this distinction, we have to see the difference between ‘but’ and ‘however’. This distinction can seem opaque since the two terms have a similar meaning. However, we now know that ‘but’ is a coordinating conjunction, which means that it is strong enough to connect two independent clauses. Other words or phrases—such as conjunctive adverbs or transitional expressions—will fail at this particular task. Knowing that only coordinating conjunctions work in this sentence pattern and being able to remember the seven coordinating conjunctions (hence the value of the mnemonic, however lame) will allow you to avoid this particular type of comma error.

Let’s look at a final example of this type of error:

The design process of a UAV starts with a pre-specified mission agenda, consequently, following conventional design methods will lead to an airplane that is only suited to achieving the primary mission.

If you find that sort of sentence in your own writing, you can do one of three things:

1. Use a semicolon.

The design process of a UAV starts with a pre-specified mission agenda; consequently, following conventional design methods will lead to an airplane that is only suited to achieving the primary mission.

This is by far the easiest and most effective solution. The semicolon probably best reflects the close relationship that the original sentence was trying to create.

2. Use a coordinating conjunction to preserve the original sentence pattern.

The design process of a UAV starts with a pre-specified mission agenda, and following conventional design methods will lead to an airplane that is only suited to achieving the primary mission.

This solution is grammatically correct but may not retain the original meaning. In this case, for instance, replacing ‘consequently’ with ‘and’ is a clear alteration of the original meaning.

3. Use a period.

The design process of a UAV starts with a pre-specified mission agenda. Consequently, following conventional design methods will lead to an airplane that is only suited to achieving the primary mission.

This is probably the least effective solution. Although it is grammatically correct, it does not express the close relationship between the two parts of the sentence. In general, replacing these sort of commas with periods can lead to unnecessarily short sentences and thus choppiness.

In sum, understanding the proper use of coordinating conjunctions will allow you to construct correct compound sentences and will allow you to avoid problems with comma use preceding conjunctive adverbs and transitional expressions.

The next comma post will look at how commas are used as a way to cope with long sentences.

* Comma use after introductory elements will be discussed in a future post.

Easing My Way into Commas

I did receive some suggestions of better titles for this post, most involving puns that I’ll leave to your imagination (think Shakespeare, Mackenzie King, 80s music, and—inevitably—Sanskrit). But since comma jokes are too easy, I am just going to stick with a boring, descriptive title.

One of the reasons that I have put off discussing commas for as long as I have is the obvious fact that there is so much to be said. In the face of such an extensive issue, it can be hard to sort out what actually needs to be said. In the classroom, I often address this problem by starting with a segment dedicated to ‘what I wish people knew about the topic at hand’. In order to prepare those remarks, I have to think hard about what I repeatedly see in student writing. Once I have figured that out, I am able to focus on discussing things that I know to be troublesome rather than on confronting the topic from all possible angles.

So what do I wish people knew about commas?

1. That understanding comma use in compound sentences—which involves understanding coordinating conjunctions—is crucial.

2. That comma use has to be more than just a response to the length of a sentence.

3. That commas come in two varieties: solitary and paired. When we use only half of a pair of commas, we create confusion.

4. That grasping how restrictive and non-restrictive elements work is essential to using commas well.

5. That comma use can be influenced by discipline.

6. That the decision about whether to use the serial comma should not be taken lightly.

I hope that these six points will work as six distinct posts. The first step, in a future post, will be to look at comma use in compound sentences. Stopping here—before getting anywhere near the nitty-gritty of commas—is wrong, I know. And I would love to dive right in, but the way commas function within compound sentences is far too important to be tucked away at the tail end of this post. If you take that as a roundabout way of saying that I just didn’t have time to write a comprehensive post on comma use in compound sentences this week, I’m okay with that.

Lastly, I wish you all a fun-filled Halloween! (Just in case you have any nagging questions about how to spell or pronounce Halloween, The Word Lady is there is clear things up!)

Colons

Today I am going to talk about colons. Before we look at the stylistic advantages of using colons, let’s look at the rules of colon use. In fact, let’s look at the only rule of colon use: a colon should appear after a complete sentence. In other words, don’t put a colon in the middle of a sentence:

This paper will consider: strategy, structure, personnel, and process.

The writer has presumably used a colon here because what follows is a list. But the decision about colon use actually needs to be made on the basis of what precedes the colon. This example is no different, logically, than the following: ‘I went to the store to buy: milk and eggs.’ In both cases, the colon is redundant because the items being listed are part of the sentence. Now consider a slight modification:

This paper will consider four areas of inquiry: strategy, structure, personnel, and process.

In this case, we have a complete sentence before the colon. If we want to go on to say what those four areas of inquiry are, we need to provide that information separately. A colon allows us to indicate that what follows will provide the information necessary to complete the thought found in the first sentence. Note that what follows a colon need not be a full sentence; single words, phrases, and full sentences can all appear after a colon.

With one simple rule—a colon should appear after a complete sentence—we are ready to see the stylistic advantages of the colon. Here are the three possible versions of our example:

This paper will consider: strategy, structure, personnel, and process.

This paper will consider four areas of inquiry: strategy, structure, personnel, and process.

This paper will consider strategy, structure, personnel, and process.

The final version seems to me to be the best. The first version, as we have seen, is incorrect, and the second version, while perfectly acceptable, seems a bit ponderous. In a sentence this simple, there is no need for a full sentence to introduce a list. (I might change my assessment of the second version if these ‘four areas of inquiry’ had ongoing structural significance for the paper; in that case, identifying them as such might be valuable.) If you find yourself putting a redundant colon in the midst of a simple list, you can just take it out and carry on. If, however, you find such a colon in a more complex sentence, it may be trying to tell you something. Consider this example:

This paper will consider: the nature of strategy in the high-tech industry in the period starting in 1983, the structure, both financial and organizational, of CHX Corporation, the management of personnel in this company at this time, and the process whereby the company came to occupy such a dominant market position.

In this case, the colon—while still technically redundant—is definitely there for a reason. The author used a colon in response to an inkling that the reader would need to know that a list is coming. So, in this case, try keeping the colon and adding the requisite words to create a full sentence before that colon.

This paper will consider four areas of inquiry: the nature of strategy in the high-tech industry in the period starting in 1983; the structure, both financial and organizational, of CHX Corporation; the management of personnel in this company at this time; and the process whereby the company came to occupy such a dominant market position.

(You will see that I have also added semicolons in the place of commas between the list items; I have done so because one of the list items had internal commas, which will potentially inhibit the readers’ ability to understand the list. I will address this issue in more detail in a future post on writing effective lists.)

I will end by commenting on the way that colons can be an efficient and elegant way to convey the relationship that exists between parts of your writing. Here is a sentence that I used at the outset of this post:

 In fact, let’s look at the only rule of colon use: a colon should appear after a complete sentence.

Here a colon allows us to fulfill the promise of the first sentence without belabouring the point. We don’t have to say ‘In fact, let’s look at the only rule of colon use, which is that a colon should appear only after a complete sentence.’ The version with the colon is, in my view, preferable since it uses fewer words and avoids a commonplace pattern of subordination.

If you have questions about colons or thoughts about the role they play in academic writing, please leave them in the comments below.

Interrupting Yourself

One of the challenging things about this blog is the process of dividing the broad topic of academic writing into manageable bits. My goal is to have posts that are short, self-sufficient, suitably comprehensive, and recognizably part of a broader whole. Since that broader whole, by definition, doesn’t exist yet, I am struggling to find the optimal way to divide things up. I recently wrote a post on dashes, which I thought would be a neat and tidy little topic. As it turned out, I actually had quite a bit to say about dashes. In the end, I ran out of space before I had dealt with the implications of interrupted sentences in academic writing (thanks to my husband for diplomatically pointing this out). So here is another attempt to discuss this issue, focussing particularly on how to decide what punctuation to use when interrupting yourself. Let’s start with that basic idea: interruption. When and how should you interrupt yourself? As I said in the earlier post, interruption is a stylistic aberration and thus not something to be overused. But there are times when we need to add extra information into our sentences. Consider the following three examples:

Civil society (a concept that is notoriously difficult to define) is crucial to any attempt to increase voter turnout in a proportional representation electoral system.

Civil society, a concept that is notoriously difficult to define, is crucial to any attempt to increase voter turnout in a proportional representation electoral system.

Civil society—a concept that is notoriously difficult to define—is crucial to any attempt to increase voter turnout in a proportional representation electoral system.

Which of these do you prefer? Chances are you prefer the second one; barring some special circumstance of which we are unaware, this particular interruption is best managed with commas. To review, commas are the least obtrusive way of interrupting yourself; parentheses announce that something is unimportant and not worthy of integration into the sentence; and dashes draw the reader’s attention to the material. In this case, the interruption should likely be handled with commas because the point needs to be made but not emphasized.

Here are some examples of sentences with unimportant interruptions (e.g., technical information, dates, terminological clarifications):

The sensor (LK-G82, KEYENCE company) is put inside a casing and then used to measure impact velocities.

Isaac Newton (1643–1727) wrote extensively on Biblical hermeneutics.

Funding was provided to elementary schools (kindergarten through grade eight) and secondary schools (grades nine through twelve).

In each of these cases, parentheses seem like the best choice; in fact, using commas in their place would require us to use extra words and would draw unnecessary attention to useful but uninteresting information.

Here, however, are some borderline instances of parentheses usage:

Individuals with unusable surveys (due to either illegibility or incompleteness) will not be included in the second level of analysis.

Earthenware vessels (among other artefacts) were present in the region.

Of these barriers to cycling, the most likely candidates include weather elements (such as rain or snow) and safety elements (such as bike lanes and bike-friendly intersections).

I would likely recommend not using parentheses in these sorts of cases. My guiding principle is that the author should have a sound reason to exclude the bracketed information from the sentence proper. In each of these three cases, I think the parentheses could be avoided by using commas.

Now let’s turn briefly back to dashes. As we saw in the earlier post, dashes indicate that something is worthy of extra attention:

The issues that confront students entering late-stream French immersion programs—most importantly, their lack of familiarity with instructional vocabularies—are a pressing concern for school administrators.

We also learned that dashes can help us when using commas would make the sentence needlessly complex. In such a case, we may decide to use dashes to interrupt ourselves even though the intervening material isn’t especially noteworthy. Consider these two examples:

Since architects are dependent on domain knowledge, generally elicited from specialists, users, and stakeholders, the development of guidelines should not be dependent on general algorithms.

Since architects are dependent on domain knowledge—generally elicited from specialists, users, and stakeholders—the development of guidelines should not be dependent on general algorithms.

The second sentence is, in my view, easier to read. However, the dashes are there for clarity rather than for emphasis, and one could quibble that I have overemphasized the interruption. It is important to find a balance; in some cases, I would definitely reword rather than use dashes to solve this type of problem.

I think I am now finished with dashes and the related question of sentence interruptions. But if you have further questions associated with this issue, please raise them in the comments below. Next we will look at an editing strategy designed to help us deal with our problem sentences.

Dashes

Today’s topic is dashes: when to use them and when not to use them. At the risk of boring you—although you are reading a post about dashes so you really have only yourself to blame—I am going to start with the distinction between hyphens and dashes. A hyphen joins two words into a unit (e.g., ‘on-site manager’ or ‘lotus-eater’) or indicates a word break at the end of a line of text. A short dash (often called an en-dash because it is the width of the letter ‘n’) separates items such as dates or page numbers (e.g., 1711–1776 or pp. 106–7). A long dash (often called an em-dash because it is the width of a letter ‘m’) is what we will be talking about below. A hyphen is produced by a simple keystroke on your keyboard; dashes are symbols that can be inserted into a document in a variety of ways. For more on hyphens and dashes, you can consult the online Chicago Manual of Style.

Just in case anyone is still reading, I am now going to talk about using dashes in your writing. First, we will look at single dashes, which I will argue are insufficiently formal for academic writing, although some would disagree. Here is an example:

Recombinant DNA technology has left an indelible mark on the basic ingredient of human survival—food production.

This dash could be replaced by either a comma or a colon. I would suggest a colon because a colon signals unambiguously that what follows will complement or complete what precedes it. A comma would work, but since commas play so many roles in our writing, I generally choose a colon over a comma when they go head-to-head. Here is another example:

The range of Hume’s writings help us to understand his reservations about a revolutionary approach—such an approach would be unlikely for a thinker so immersed in historical considerations.

In this case, you could replace the dash with either a semicolon or a period. Generally, a semicolon will work best since it will provide the closeness that the author was trying to convey with the dash.

I am only arguing against the single dash in academic writing—it is absolutely fine in a more casual context like, say, a blog post. My reason for this advice is that the dash can play the role of a comma, colon, or period without doing anything unique. Nothing special is added and something may be lost. You may lose clarity, and you will certainly lose formality. I am being, of course, overly opinionated; if you wish to defend the use of single dashes in formal writing, I hope you’ll do so in the comments below. One possible line of defence is certainly the drama of the single dash; since it breaks up a sentence in an unconventional manner, it does draw real attention to what follows it. Now let’s turn to the use of double dashes.

Double dashes are used instead of commas (or parentheses) to interrupt a sentence. The phrase separated by dashes must be grammatically inessential, by which I simply mean that the sentence will still work without that phrase. Here is an example:

The issues that confront students entering late-stream French immersion programs—most importantly, their lack of familiarity with instructional vocabularies—are a pressing concern for school administrators.

It is often said that we use dashes to signal that something crucial is being added to a sentence and use parentheses to signal that the interruption is relatively unimportant (e.g., to give dates or citations or examples). In this schema, commas fall somewhere in the middle. There is some truth to this division, but I find the next three principles more helpful in deciding whether to use double dashes:

1. Most importantly, double dashes can be used to add an element to a sentence that already has its share of commas.

2. Double dashes are very common in some fields (most humanities fields, for instance) and rare in others (in many scientific fields, for instance). If they are rare in your field, it is unlikely that you would want to use them.

3. Regardless of field, double dashes are a stylistic variant and should be used sparingly; they draw attention to themselves, and the audience will grow tired of them if they are overused.

Overall, I suggest using double dashes—if they are appropriate to your field of study—in moderation and replacing single dashes with colons, commas, or even periods, if only in your formal writing.

Note: For more on dashes, see Interrupting Yourself.