# Can three people sit in a circle?

Is it appropriate to say:

The three characters sat down in a circle.

When it means that three people sat down facing each other? Technically this formation would be a triangle, but is it appropriate usage to say that they sat in a circle, as with a larger group?

I am a student and wrote a similar sentence in a creative response draft and had it returned with the word 'circle' crossed out and 'triangle' written in its place. However, I feel my usage makes perfect sense and that correcting it to 'triangle' is a bit pedantic.

Does my word usage make sense and seem natural or should it be corrected?

• The point is to get your message across, which your original sentence does perfectly. The change to triangle is pedantic, just as you said. Furthermore, sitting in a circle can easily be inferred as sitting down facing each other, which was your original intent. – vickyace Mar 13 '17 at 6:29
• This is clearly a question about Euclidean geometry and not English usage. Where are the mods? – deadrat Mar 13 '17 at 6:49
• GWH, you may ignore my previous comment. Although three points determine a circle, three people are probably not sufficient for your purposes. The problem with such a small number is that your statement leads to an ambiguity, namely that the three sit within a circular area. – deadrat Mar 13 '17 at 6:52
• @deadrat And just how many people are sufficient to define a circle? Five? Wait, that's s pentagon. Six? Hmm, still no, a hexagon. And seven is right out... Ok, let's just skip to 1,000 people. No, wait, that's a chiliagon. So it looks like people can't sit in a circle, unless there was a circle painted on the ground to begin with. – Dan Bron Mar 13 '17 at 9:09
• The other aspect to consider, besides the bare geometry, is that nobody sits in triangles, and thus the phrase isn't idiomatic, and would call attention to itself. Why are they sitting in a triangle? What kind? A right triangle? Isosceles? Which way does it point? Is that significant? etc. A circle is just a circle, that's how people sit. Language is not math. My phone "rings" despite having no bell, and I end calls by "hanging up" despite there being no switchhook. People sit in circles even if they also happen to be sitting in some other polygon described by the number of people sitting. – Mr. Shiny and New 安宇 Mar 13 '17 at 15:05

Yes.

Mathematically, three points in space define a unique circle, so I think OP has metaphorical, mathematical and common-sense arguments that the characters can sit in a circle not just a triangle.

https://math.stackexchange.com/questions/213658/get-the-equation-of-a-circle-when-given-3-points

• Come for the English language & usage, stay for the math! I read deadrat's "three points determine a circle" above and was ignorantly-incredulous (single word request?). Sincerely, thanks for this link. And now on topic, I fully agree that OP's usage is justified on many fronts. – MDHunter Mar 13 '17 at 12:20
• Ok, now figure out the math of how I can upvote you twice! – Dan Bron Mar 13 '17 at 12:59
• Mathematically the statement, three people sat down in a circle, and the statement, three people sat down not in a straight line, say the same thing. But as a matter of plain English, these statements certainly do not say the same thing. I find the "correction" comically wrongheaded. – Airymouse Mar 13 '17 at 13:38
• Are these people sufficiently thin that they may be considered as points? – davidlol Mar 13 '17 at 20:11
• @HotLicks Yes agreed. It was the OP's real-world reviewer who mentioned triangles, and the point of my answer is that even from a maths perspective circle is ok. – k1eran Mar 14 '17 at 18:38

Circle is used metaphorically to suggest a way of organizing things with no specific point being special with respect to the other points.

On a circle, all points are equal - they are equidistant from the center.

The Paris Peace Accords demonstrates this metaphor, and at the same time illustrates that an object with straight edges is a signal of conflict:

A similar debate concerned the shape of the table to be used at the conference. The North favored a circular table, in which all parties, including NLF representatives, would appear to be "equal"' in importance. The South Vietnamese argued that only a rectangular table was acceptable, for only a rectangle could show two distinct sides to the conflict.

On a circle, there is room for more, all with the same attributes of any other point. On a polygon of given number of sides, there is no room for more.

If you were to change circle to triangle, this equality would be overlooked, unless you specified equilateral triangle. This is so much more ungainly than circle that it rolls off the tongue like a spiky thorn. A triangle by itself grates almost as much.

Imagine the leader of a small wagon train (from the American West) telling the drivers to put the wagons in a triangle! It doesn't matter that there might be only three wagons - he tells them to circle the wagons!

Even when you sit at the campfire, you sit around the fire, as "in circumference, in a circle, on every side" (Etymonline.com). You can do this even if you are alone, even though, at any particular moment, you can't subtend an arc of more than a few degrees without being scorched.

Circular logic is not circular, but sometimes takes a wavering path of confusion before reaching its starting point. Yet it's still called circular.

I think you did fine using circle.

Three people sat facing each other, in a circle of sorts.

This is probably how I would write it.Having said that, there is nothing wrong with your description, "a circle of three", especially in a creative writing assignment.

Your guide apparently needs to widen his/her horizons.

Just to make a case in point there is even a novel titled "Circle of Three" by Rohit Gore [http://www.goodreads.com/book/show/15990555-circle-of-three][1]

• OK... but the title of the book is probably metaphorical, I would not interpret it as meaning the three people formed a circle when they sat down together. – Mari-Lou A Mar 13 '17 at 8:27
• The term is not being used in the geometric sense! It is more metaphorical. "Circle of sorts" is not appropriate -- it just creates a distraction. – Hot Licks Mar 14 '17 at 18:23

Merriam Webster defines:

### Circle

a perfectly round shape

a line that is curved so that its ends meet and every point on the line is the same distance from the center

a path that goes around a central point

an arrangement of people or things that forms a circle

According the third sense here, it would seem that as long as there's a central or common point of reference (including in a figurative sense), three people are quite able to sit in a circle.

So it looks like people can't sit in a circle,....

Ha ha ha ha ha. The hilarity never ends on ELU. You see, it's funny because, well, never mind.

Obviously what's required is the development of a theory of circle sitting or circumsituation, if you will.

Axiom 1: Two people can't sit in a circle.

Now, for this metaphor to make sense, the circle must be of sufficient diameter:

Axiom 2: For people to sit in a circle, the circle must have a minimum radius r.

If a circle is too small, then the people will just be seen as a bunch or crowd. We'll leave r unspecified, but for purposes of estimation, let's say that r = A*m/2*π, where m is the minimum number of circumsituents anyone could propose, and A represents a reasonable arm's length between circumsituents.

Theorem 1: There is some number of people who can sit in a circle. Proof: Take a circle of radius at least r and add people to the circumference until each person's shoulders touch another person.

Definition 1. An even number of people are sitting in a circle C of radius at least r, if every circumsituent sits on the circumference of C equidistant in arc length on C from his neighbors, and his line of sight through the center of C includes another circumsituent.

Definition 2. An odd number N of people are sitting in a circle C of radius at least r, if N-1 people can sit in the circle C.

Theorem 2: 3 people cannot sit in a circle. The proof is left as an exercise for the reader. Hint: Apply Definition 2.

• I contest the final requirement in your Def 1 and its attendant consequence of creating a need for Def 2. I can construct definitions which permit odd-numbered circles directly, without induction from even-numbered circles. – Dan Bron Mar 14 '17 at 18:04