Should there be spaces around —?

It depends on the style guide more than the country. Chicago and most book publishers set the em dash closed up, with no space on either side. AP style, which many US newspapers follow, puts one space before and one after. British publishers usually skip the em dash altogether and use a spaced en dash in its place.

Does using — mean a text was written by AI?

No. The em dash is ordinary punctuation that writers used for centuries before chatbots existed, and plenty of careful human writers use it every day. It became a talking point because chatbots reach for it often, so some readers now treat it as a tell. If you like the mark, keep it, and mix in commas and parentheses so it doesn't land in every sentence.

Why do Spanish novels use — instead of quotation marks?

Spanish marks dialogue with the raya, a dash the same length as the em dash. A line of speech opens with it, attached to the first word, and another pair of dashes brackets the narrator's words, such as dijo ella, when they interrupt the speech. French and Russian fiction often use a dash for dialogue too, so it isn't a translation slip.

Why is più written as pi— in my old Italian file?

The dash is the letter ù. DOS stores ù at byte 151, and Windows-1252 has the em dash there, so the ù at the end of più and giù shows up as a long dash, and French où gets the same treatment. Loading it into MySQL, add CHARACTER SET cp850 to the LOAD DATA INFILE statement so the letters come back.

ASCII 151

Em dashLowercase u with grave accent

ASCII 151 is — on Windows and ù in old DOS.

— (Em Dash) is byte 151 in Windows-1252, Unicode U+2014. It marks a break in a sentence. American style sets it with no spaces on either side, while the shorter en dash (byte 150) handles ranges, and plain-text email and typewriters use two hyphens as a stand-in. One layout surprise: line-breaking rules allow a break both before and after an em dash, so an unspaced one can start a line on its own or split a phrase you meant to keep together. Wrap the phrase in white-space: nowrap, or put a word joiner (U+2060) next to the dash, if that matters. When its UTF-8 bytes E2 80 94 get read as Windows-1252 they turn into â€”, and that last character is a closing curly quote, so the damage looks like a stray quotation. The HTML entity is —.

ù (Small U with Grave) is byte 151 in code page 437, the original IBM PC character set, and maps to Unicode U+00F9. Windows-1252 has the em dash — at 151, so Italian DOS text read as Windows-1252 turns più into pi—, and French où into o—. Code page 850 kept ù at 151 too.

PunctuationLatin Letter

Technical Details

Code Example

</>

#include <stdio.h>

int main(void) {
    /* On Windows (code page 1252) byte 151 is —.
       Modern terminals expect UTF-8, so print the code point, not the byte. */
    printf("\u2014\n");   /* UTF-8: E2 80 94 */

    unsigned char b = 151;         /* the raw Windows-1252 byte */
    printf("%d 0x%02X\n", b, b);    /* 151 0x97 */
    return 0;
}

Code Example

</>

#include <stdio.h>

int main(void) {
    /* Byte 151 is ù only on a console using code page 437
       (chcp 437 on Windows, or DOSBox). On a UTF-8 terminal the
       lone byte 0x97 is invalid and prints as garbage, often �. */
    putchar(151);

    /* char is signed on x86, so a plain char holding 0x97 is -105.
       Use unsigned char when you compare or index by byte value. */
    char c = (char)151;
    unsigned char u = 151;
    printf("\n%d %d\n", c, u);   /* -105 151 */

    /* Portable: print the Unicode character as UTF-8 instead. */
    printf("\u00F9\n");          /* C3 B9 */
    return 0;
}

How to Type It

Windows: Hold Alt and type 0151 on the numeric keypad, with Num Lock on. The leading zero picks the Windows code page. No numeric keypad? In Word, type 2014 and press Alt+X.
Mac: Press Option+Shift+-.
Linux: In GNOME, or any desktop running IBus, press Ctrl+Shift+U, type 2014, then Space. With a Compose key set up, press Compose, minus, minus, minus.
Phone: iPhone: open the 123 keyboard, long-press the hyphen key and pick —. Android (Gboard): tap ?123, long-press the hyphen key and pick —.

How to Type It

Windows: Hold Alt and type 151 on the numeric keypad, with no leading zero. Windows reads it as a DOS code and inserts ù.

ù is also ASCII 249 on Windows, which has the full guide.

Frequently Asked Questions

It depends on the style guide more than the country. Chicago and most book publishers set the em dash closed up, with no space on either side. AP style, which many US newspapers follow, puts one space before and one after. British publishers usually skip the em dash altogether and use a spaced en dash in its place.
No. The em dash is ordinary punctuation that writers used for centuries before chatbots existed, and plenty of careful human writers use it every day. It became a talking point because chatbots reach for it often, so some readers now treat it as a tell. If you like the mark, keep it, and mix in commas and parentheses so it doesn't land in every sentence.
Spanish marks dialogue with the raya, a dash the same length as the em dash. A line of speech opens with it, attached to the first word, and another pair of dashes brackets the narrator's words, such as dijo ella, when they interrupt the speech. French and Russian fiction often use a dash for dialogue too, so it isn't a translation slip.
The dash is the letter ù. DOS stores ù at byte 151, and Windows-1252 has the em dash there, so the ù at the end of più and giù shows up as a long dash, and French où gets the same treatment. Loading it into MySQL, add CHARACTER SET cp850 to the LOAD DATA INFILE statement so the letters come back.

Frequently Asked Questions

The dash is the letter ù. DOS stores ù at byte 151, and Windows-1252 has the em dash there, so the ù at the end of più and giù shows up as a long dash, and French où gets the same treatment. Loading it into MySQL, add CHARACTER SET cp850 to the LOAD DATA INFILE statement so the letters come back.

ASCII 151

Em dashLowercase u with grave accent

ASCII 151 is — on Windows and ù in old DOS.

Punctuation

Latin Letter

#include <stdio.h>

int main(void) {

/* On Windows (code page 1252) byte 151 is —.

Modern terminals expect UTF-8, so print the code point, not the byte. */

printf("\u2014\n"); /* UTF-8: E2 80 94 */

unsigned char b = 151; /* the raw Windows-1252 byte */

printf("%d 0x%02X\n", b, b); /* 151 0x97 */

return 0;

}

#include <stdio.h>

int main(void) {

/* Byte 151 is ù only on a console using code page 437

(chcp 437 on Windows, or DOSBox). On a UTF-8 terminal the

lone byte 0x97 is invalid and prints as garbage, often �. */

putchar(151);

/* char is signed on x86, so a plain char holding 0x97 is -105.

Use unsigned char when you compare or index by byte value. */

char c = (char)151;

unsigned char u = 151;

printf("\n%d %d\n", c, u); /* -105 151 */

/* Portable: print the Unicode character as UTF-8 instead. */

printf("\u00F9\n"); /* C3 B9 */

return 0;

}