About Double Question Mark
Double Question Mark is the Unicode character ⁇ at codepoint U+2047 in the General Punctuation block of the Basic Multilingual Plane (BMP).
Reference
Two question marks in one cell, encoded for vertical East Asian text. Unicode files it as Po in General Punctuation with the compatibility decomposition 003F 003F, so NFKC and NFKD pull it back into two ordinary question marks. It carries Terminal_Punctuation and Sentence_Terminal and has line-break class NS, nonstarter. It arrived in Unicode 3.2 in 2002, later than the three marks it is usually grouped with, and unlike two of them it has no emoji property at all — so it renders as text everywhere, with no coloured form available. There is no HTML named entity and no Windows-1252 slot.
Typical uses
- A doubled question mark in one cell of vertical Japanese text
- Any context where a doubled mark must be a single character to kern, centre or style
- Compatibility with legacy East Asian character sets
At a glance
- U+2047, General Punctuation; Unicode 3.2 (2002); general category Po
- Terminal_Punctuation and Sentence_Terminal; line break class NS
- No emoji property of any kind — always text
- No HTML named entity; no Windows-1252 slot; NFKC gives 003F 003F
Historical Notes
Origins
It belongs to the small group of doubled punctuation characters encoded for vertical setting, where a mark has to fit one square cell of the grid.
Unicode
Unicode 3.2, released in 2002 — nine years after the double exclamation mark and three after the two mixed marks, which makes this the last of the four to be encoded. The chart entry carries no alias, no annotation and no cross-reference; only the decomposition line, which states that the character is two question marks.
The emoji divide
Four characters pack a doubled question or exclamation mark into a single cell: the double exclamation mark, the exclamation question mark, the question exclamation mark and this one. Two of the four carry emoji properties and two do not. This one does not, and neither does the question exclamation mark. The reason is which marks happened to be in the Japanese carrier sets when those were mapped into Unicode, and the practical result is that this character keeps the weight and colour of the surrounding text under all circumstances, where its emoji-bearing siblings can be given a coloured form by a variation selector.
Developer Notes
| Unicode Codepoint | |
| Decimal | |
| UTF-8 (hex) | |
| UTF-16 (hex) | |
| HEX Entity | |
| HTML Entity | |
| CSS (\xxxx) | |
| JavaScript | |
| Python |
How to Use Double Question Mark
To use the Double Question Mark symbol (⁇) in HTML, insert the
entity ⁇ directly into your markup. The decimal entity ⁇ works in any HTML-serialised document.
In CSS pseudo-elements use content: '\2047';.
In JavaScript and TypeScript the escape '\u2047' evaluates to ⁇. In Python source, write '\u2047'.
The character encodes to E2 81 87 as UTF-8 bytes and 2047 as UTF-16 code units. Its decimal codepoint is 8263.
This is a text-based Unicode character. It will render in the current font on most platforms. If a font lacks this glyph, the browser may show a placeholder or fall back to a system font.
Things to know
- NFKC and NFKD decompose it to two ordinary question marks, so a normalising pipeline will not preserve the single character.
- No emoji property at all, so it can never render as a coloured glyph. That is a feature in running prose.
- Line break class NS: a line will not begin with it.
- Terminal_Punctuation and Sentence_Terminal both apply.
- No named HTML entity — write ⁇ or ⁇ — and no Windows-1252 slot. Font coverage is worth testing.
In the Wild
On 5 August 2026 the character appeared twice on the Japanese Wikipedia article about punctuation marks, both times as a specimen.
The page worth reporting is one that did not contain it at all. English Wikipedia's article on chess annotation symbols runs to 22,500 characters and is largely about doubled punctuation — the double question mark for a blunder, the question mark and exclamation mark in combination for a dubious or an interesting move. It carried fifty-two ordinary question marks and fifty-six ordinary exclamation marks, and not one of these single-character forms.
That is the strongest available test of the argument for the character, and the page fails it: the one field routinely cited as needing a single-glyph doubled question mark writes two ASCII characters instead, on the encyclopaedia page devoted to its own notation. Pages change; this is what those showed on the day.
FAQ
Is it different from typing ??
Formally it decomposes to exactly that, so normalisation makes them identical. The difference is that this is one character to style, kern or centre.
Will it turn into an emoji?
No. Unlike the double exclamation mark and the exclamation question mark, it carries no emoji property at all.
Is there an HTML entity?
No named one. Use ⁇ or ⁇.
When was it encoded?
Unicode 3.2, in 2002 — the last of the four doubled punctuation marks in its group to be added.