Bracket and Parenthesis Symbols

Brackets and parentheses are among the most fundamental punctuation marks in human writing, yet most people only know the handful available on a standard keyboard. Unicode assigns well over a hundred distinct bracket characters spanning mathematical notation, CJK typesetting, ornamental publishing, and specialized technical domains. This guide catalogs every major bracket and parenthesis character in Unicode, explains how they are classified, and provides code points and practical usage notes for developers, designers, and writers.

Why So Many Brackets?

The ASCII character set offers exactly six bracket characters: ( ), [ ], and { }. These suffice for English prose and basic programming, but the world's writing systems and technical notations demand far more variety. Chinese and Japanese typesetting use fullwidth and corner brackets. Mathematics requires angle brackets, ceiling and floor brackets, and double brackets. Medieval manuscripts used ornamental brackets for marginalia. Unicode encodes all of these as distinct characters with precise semantics.

Understanding which bracket to use matters for correctness. A mathematical inner product written with less-than and greater-than signs (<x, y>) is semantically wrong — the correct characters are mathematical angle brackets (⟨x, y⟩). Search engines, screen readers, and parsers can all distinguish between these, so choosing the right bracket improves accessibility, searchability, and interoperability.

ASCII Brackets

The six ASCII brackets remain the most widely used:

These six code points are universally supported across every font, terminal, and programming language. They are the foundation upon which all other bracket characters build.

Mathematical Brackets

Mathematics uses specialized brackets that should not be confused with ASCII look-alikes:

A common mistake is using < (U+003C, LESS-THAN SIGN) and > (U+003E, GREATER-THAN SIGN) as angle brackets. These are comparison operators, not brackets. In LaTeX, the correct commands are \langle and \rangle, which produce U+27E8 and U+27E9.

In Python, you can verify the distinction:

import unicodedata

# These are comparison operators, NOT brackets
print(unicodedata.name(chr(0x003C)))  # LESS-THAN SIGN
print(unicodedata.name(chr(0x003E)))  # GREATER-THAN SIGN

# These are the correct mathematical angle brackets
print(unicodedata.name(chr(0x27E8)))  # MATHEMATICAL LEFT ANGLE BRACKET
print(unicodedata.name(chr(0x27E9)))  # MATHEMATICAL RIGHT ANGLE BRACKET

CJK Brackets

Chinese, Japanese, and Korean typesetting uses a rich set of brackets not found in Western writing. Many of these are fullwidth (occupying the same width as a CJK ideograph) to maintain grid alignment:

Japanese uses corner brackets 「」 as primary quotation marks (equivalent to English double quotes) and white corner brackets 『』 for nested quotations or book titles. Chinese uses double angle brackets 《》 for book and film titles. These are not interchangeable with Western angle brackets or guillemets.

Ornamental and Typographic Brackets

Publishing and decorative typography use additional bracket styles:

These ornamental brackets appear primarily in typeset documents, invitations, and decorative contexts. Font support varies, so always test rendering before using them in production.

The Bidi Mirroring Property

Unicode brackets have a special property called Bidi_Mirrored. In right-to-left text (Arabic, Hebrew), opening brackets appear on the right and closing brackets on the left. The rendering engine automatically mirrors bracket glyphs in RTL context.

You can check this property programmatically:

import unicodedata

# Check if a character is bidi-mirrored
print(unicodedata.mirrored(chr(0x0028)))  # 1 (True) - LEFT PARENTHESIS
print(unicodedata.mirrored(chr(0x0041)))  # 0 (False) - LATIN CAPITAL LETTER A

This means that (hello) in an RTL paragraph renders with the parentheses visually flipped — the ( appears on the right side and ) on the left, matching the reading direction.

Bracket Categories in Unicode

Unicode classifies bracket characters using the General Category property:

The Ps/Pe distinction is critical for text processing algorithms. Unicode's bidirectional algorithm, line breaking algorithm, and bracket-matching utilities all rely on these categories to pair brackets correctly.

Using Brackets in HTML

In HTML, brackets can be entered directly as Unicode characters or via named/numeric entities:

Common Mistakes

1. Using <> as angle brackets. The less-than and greater-than signs are comparison operators. Use U+27E8/U+27E9 for mathematical angle brackets or U+3008/U+3009 for CJK angle brackets.

2. Mixing fullwidth and halfwidth brackets. In CJK text, using halfwidth () next to fullwidth characters creates uneven spacing. Use fullwidth （） for consistent grid alignment.

3. Ignoring RTL mirroring. Hardcoding bracket direction (e.g., manually placing ) on the left in Arabic text) breaks the bidirectional algorithm. Let the renderer handle mirroring automatically.

4. Confusing guillemets with angle brackets. The characters «» (U+00AB/U+00BB) are quotation marks used in French, Russian, and other languages. They are not brackets.

Summary

Unicode provides a comprehensive taxonomy of brackets covering ASCII, mathematics, CJK typesetting, and ornamental design. Choosing the semantically correct bracket character improves text processing, accessibility, and visual quality. Developers working with multilingual text should be especially aware of the bidi mirroring property and the distinction between visually similar but semantically different bracket characters.