Many punctuation marks are distinguished by only a stroke, a curve, or a little extra length. Their names are useful because they describe function as well as appearance. The distinctions below help when editing prose, preparing data, searching Unicode, or explaining exactly which character a document needs. Conventions vary by language and house style, so these are practical descriptions rather than universal commandments.
Hyphens, dashes, and the minus sign
| Mark | Name | Code point | Representative use |
|---|---|---|---|
| - | hyphen-minus | U+002D | plain-text stand-in; source code |
| ‐ | hyphen | U+2010 | well-being |
| – | en dash | U+2013 | pages 18–24 |
| — | em dash | U+2014 | a break—or interruption—in prose |
| − | minus sign | U+2212 | 9 − 4 = 5 |
A hyphen joins compound words or divides a word at a line ending. The common keyboard character is formally the hyphen-minus because older encodings made one key serve both hyphenation and subtraction. Unicode provides a dedicated hyphen, but U+002D remains normal in URLs, filenames, and code.
An en dash often marks a span (“Monday–Friday”) or a relationship (“the Portland–Seattle route”). An em dash marks a stronger sentence break. Spacing around em dashes depends on editorial style. The minus sign is a mathematical operator, aligned and spaced for equations. A negative temperature is properly −5 °C, though software inputs may require the ASCII hyphen-minus.
Apostrophes, quotation marks, and primes
The apostrophe and closing single quotation mark often share the same curly character: ’ (U+2019). An apostrophe marks possession or omitted material: “the editor’s note” and “don’t.” A leading apostrophe in a shortened year should curl like an omitted character: “the class of ’26.” The straight mark ' (U+0027) is an apostrophe in plain text and a delimiter in many programming languages.
English typography commonly uses left and right double quotation marks “ ” (U+201C and U+201D), and matching single forms ‘ ’ (U+2018 and U+2019). Their placement and nesting rules differ across languages and styles. Straight quotation marks are safest in source code; curly quotation marks are designed for prose.
Prime marks are not quotation marks. The prime ′ (U+2032) and double prime ″ (U+2033) represent units such as feet and inches, minutes and seconds of arc, or derivatives in mathematics. Write 5′ 10″ for five feet ten inches when the context is clear, not 5’ 10”. Time durations usually read more accessibly as “5 min 10 s.”
Dots: period, ellipsis, bullet, and middle dot
An ellipsis represents omitted material or a trailing thought. Unicode supplies the single character … (U+2026); three periods ... are also widely used. Editorial styles disagree about surrounding spaces and about how an ellipsis interacts with a sentence-ending period. Consistency and clarity matter more than choosing one encoding in every medium.
A bullet • (U+2022) is a list or separator mark. In semantic HTML or Markdown, create a real list rather than typing bullets into a paragraph; assistive technology can then announce its structure. A middle dot · (U+00B7), also called an interpunct in some contexts, can separate compact items, indicate multiplication in some mathematical styles, or perform language-specific orthographic work. It is not simply a smaller bullet.
Slash and solidus
The familiar forward slash / (U+002F) has the Unicode name solidus and the everyday name slash. It appears in URLs, fractions written linearly, alternatives, dates, and line divisions. Those uses are not interchangeable: “and/or” compresses a logical choice, while https:// follows a technical grammar. In continuous prose, words are often clearer than a dense chain of slashes.
The backslash (U+005C) is a different character used in computing and some mathematical notation. Calling it “slash” without qualification invites errors when dictating a path or command.
Section sign, pilcrow, and reference marks
The section sign § (U+00A7) points to a numbered section, especially in legal and scholarly references: § 12 or §§ 12–14. Whether a space follows it depends on style, but it should stay with its number when possible.
The pilcrow or paragraph mark ¶ (U+00B6) denotes a paragraph. Editors use it in references, and word processors display it as a nonprinting-character indicator. It need not be inserted at the end of ordinary web paragraphs; HTML already records paragraph structure.
An ampersand & (U+0026) is a logogram for “and.” It is conventional in names, compact labels, and citations, but prose styles often prefer the word. In HTML source, a literal ampersand begins a character reference and is normally written &.
Choosing the right character
Start with meaning. Is the mark joining a word, expressing a numeric range, subtracting a value, indicating measurement, or marking a quotation? Then consider the medium. A publishing system can preserve typographic Unicode characters; a programming language, identifier, or data interchange format may deliberately require ASCII.
Copying safely: the specimen boxes in this article are selectable. When exact identity matters, record the Unicode code point as well as the visible character. Similar glyphs can be indistinguishable in a particular font.
Automatic “smart punctuation” is helpful but fallible. It can turn feet into quotes, a leading apostrophe into an opening quote, or a programming delimiter into a curly mark. Review conversions in measurements, code, proper names, and copied data. Typography supports meaning best when the character is correct before its appearance is refined.