Roman means upright, non-italic type — the default style of a typeface. The name traces to ancient Rome: the monumental carved inscriptions on buildings like Trajan’s Column (113 CE) gave humanist scholars and printers of 15th-century Italy their model for reviving classical letterforms. Italian printers called this style roman to distinguish it from the Gothic (Blackletter) type that dominated German printing at the time. The name stuck, and now every normal upright font is technically “roman.”

Kerning is the adjustment of space between two specific characters. Different pairs create different optical gaps — VA looks too open, ff can look too tight. Kerning corrects these pair by pair. The word comes from the kern: the part of a metal type character that overhangs the adjacent piece.

Tracking (also called letter-spacing) is uniform spacing applied equally across a range of characters — a word, a line, or an entire text block. Used for stylistic effect or readability: tight on large headlines, looser for small caps or airy layouts. Unlike kerning, tracking doesn’t know about specific pairs — it just shifts everything evenly.

In the age of metal type, rows of text were set in rigid frames called galleys. To increase the space between lines, compositors inserted thin strips of lead (the metal — עופרת) between the rows. These strips were called leads, and the technique became known as leading. When digital typesetting arrived, the term came with it. Today’s “leading” setting in any design application is named after a physical strip of metal that no one has touched in decades.

Italic is a separately designed typeface, not just a slanted version of the roman. It originated in Venice around 1500, when Aldus Manutius commissioned a style based on the flowing handwriting of Renaissance humanist scholars. True italics have distinct letterforms — notice how the a and f look completely different. Named after Italy.

Oblique is a mechanically or optically slanted version of the roman. The letterforms are essentially the same, just tilted — no separate design. Common in geometric sans-serifs like Futura or Helvetica, where a “true” italic would look out of place.

Slanted is an informal term, often used interchangeably with oblique. Not a formal classification — more of a description.

Cursive technically refers to letterforms with connected or flowing strokes, like handwriting. Sometimes used as a synonym for italic, especially in CSS.

Script is a typeface that mimics handwriting or calligraphy, often with letters that visually connect. More decorative and extreme than cursive — think wedding invitations, not body text.

Condensed is a purposefully designed narrower variant of a typeface family. The spacing, proportions, and details are all reworked for the narrower width — not mechanically squished. Standard condensed widths are typically around 75% of the regular width.

Compressed is a more extreme degree of narrowing than condensed. Some families offer a full progression: Regular → Condensed → Extra Condensed → Compressed.

Narrow is essentially a synonym for condensed, used in some typeface names (Arial Narrow, for example). No strict distinction from condensed.

A note on bad practice: mechanically scaling a font horizontally to make it narrower is called horizontal scaling or distortion. It creates ugly, uneven strokes and is generally considered a design mistake — use a real condensed font instead.

Extended is a purposefully designed wider version of a typeface. Like condensed, it’s a genuine redesign for greater width, not mechanical stretching.

Expanded is a direct synonym for extended — used interchangeably in font naming. Some families use one, some use the other.

Wide is an informal or marketing term sometimes used as a font style name, equivalent to extended.

A note on bad practice: horizontal scaling a font to make it wider produces distorted, uneven strokes. Always use a designed extended variant when available.

In justified text, software stretches word-spacing to fill each line edge to edge. When the gaps between words happen to align visually across multiple consecutive lines, they form a diagonal or vertical channel of white space running through the paragraph — called a river.

Rivers are especially common in narrow columns, when words are long, or when automatic hyphenation is turned off. They distract the eye and break the flow of reading. Solutions: enable hyphenation, use ragged-right alignment, widen the column, or manually edit line breaks.

An orphan is a single line from the beginning of a paragraph left isolated at the bottom of a column or page, cut off from the lines that follow on the next page.

A widow is a very short line — often just one word — left alone at the top of a new column or page, separated from the rest of the paragraph above.

Both are considered typographic flaws because they look unfinished and interrupt the reading rhythm. Fix them by adjusting tracking, editing the text, or using paragraph’s widow/orphan control settings.

Note: Some style guides reverse the definitions of widow and orphan. The key principle is the same: isolated lines at boundaries should be avoided.

An initial is the general term for a decorative or enlarged first letter at the start of a text — a tradition that goes back to illuminated manuscripts.

A drop cap is an initial that drops down into the text below it, with the surrounding lines wrapping around it. The enlarged letter’s top aligns with the first line, and its bottom sits several lines down into the paragraph.

A raised cap (or versal) is an enlarged initial that sits on the baseline and rises above the text, without the surrounding lines wrapping around it.

Drop caps are used to create a strong visual entry point into a long text — common in books, magazines, and editorial design.

A paragraph indent is the small horizontal space at the start of a line that signals “a new thought begins here.” Its origin is practical: in medieval manuscripts, scribes left a blank space at the start of each paragraph for a decorated initial letter — a large, ornate capital drawn by a specialist called a rubricator. When the rubricator didn’t show up (or the book was left unfinished), the gap stayed — and the convention stuck.

After Gutenberg, printers inherited this habit and kept it even without any initial letter. Five centuries later, the indent is still the default signal for a new paragraph in books and print.

Two conventions to know:
In traditional book typography, do not indent the first paragraph after a heading — the heading already signals a new section, so the indent is redundant. Indent only the second paragraph onward.

An alternative convention, common in European and screen typography, uses space between paragraphs instead of an indent — and no indent at all. This is the default in HTML. Using both indent and space is generally considered redundant.

All four words describe typefaces without serifs — but they come from different countries and centuries, and each carries slightly different design connotations.

Sans-serif (or simply sans) is French for “without serif.” The standard neutral term in English and international design contexts today.

Grotesque (and its German spelling Grotesk) was the original name for sans-serif type when it first appeared in the early 19th century. The style was considered strange and ugly by contemporary standards — hence “grotesque.” Today it names a specific subcategory: the early, slightly irregular sans-serifs of the 19th and early 20th century — Akzidenz-Grotesk, Helvetica, Univers.

Gothic is the American term for sans-serif, rooted in newspaper and sign-painting traditions: Franklin Gothic, News Gothic, Trade Gothic. This creates a real confusion: in European typographic tradition, Gothic means Blackletter (Fraktur, Textura) — the dense medieval script that is essentially the opposite of a clean sans-serif. The same word, two incompatible meanings, depending on which continent you’re on.

The website presents two poles of type history: the Humanist tradition (Renaissance, broad-nib pen, diagonal stress, low contrast) and the Modern tradition (Bodoni, Didot, pointed nib, vertical stress, extreme contrast). Between them sits a third era that doesn’t appear in the site: Transitional type.

Transitional typefaces — roughly 1720–1800, associated with designers like John Baskerville and William Caslon — sit exactly in the middle. They show more vertical stress than humanist type, greater contrast between thick and thin strokes, and more refined serifs — but they haven’t yet reached the extreme geometry of Bodoni or Didot. The shift reflects the replacement of the quill with the pointed metal nib as the model for letterform design.

Key typefaces: Baskerville, Times New Roman, Caslon, Century.

A note on the name: “Transitional” is an awkward label — it defines a style by what came before and after it, not by its own character. Some scholars prefer “Baroque” or “Realist” instead.

Every stroke in a letterform originates in a specific pen movement. Understanding this explains why letters look the way they do.

Downstroke: pulling the pen downward, toward the writer. With a broad-nib pen held at a consistent angle, the nib’s full width meets the paper — producing a thick stroke.

Upstroke: pushing the pen upward or to the right. The nib’s narrow edge meets the paper — producing a thin stroke.

The ratio between the two is called stroke contrast. High contrast (thick/thin difference is extreme) = pointed-nib or Didone style (Bodoni, Didot). Low contrast (nearly uniform strokes) = sans-serif or monoline style (Futura, Helvetica).

The angle at which thick and thin transitions happen is called the stress axis. A diagonal axis (~30–45°) reflects the natural hold of a broad-nib pen — this is humanist type. A vertical axis reflects a more mechanical, upright approach — this is modern/Didone type.

Before footnote numbers, printers used a standardized sequence of symbols to mark references and annotations. In order:

* Asterisk — the first footnote mark. From Greek asteriskos (little star). Also used for emphasis, censorship of letters, and wildcards in computing.

† Dagger — second footnote mark. Also placed after a person’s name to indicate they are deceased. Called obelisk (Greek: small spit or spike).

‡ Double Dagger — third footnote mark. Also called diesis or double obelisk.

§ Section Sign — marks a legal or legislative section (§ 4.2). Standard in legal writing worldwide.

¶ Pilcrow — the paragraph mark. Originally used to signal a new section of continuous text, before the convention of starting a new line existed. Today it appears in word processors to show invisible paragraph breaks, and survives in legal writing.

& Ampersand — a ligature of the Latin word et (“and”). The name derives from the phrase “and per se and” — meaning “&, by itself, means and” — which was recited at the end of the alphabet in English schools.

‽ Interrobang — a combined question mark and exclamation point, proposed in 1962 by American ad executive Martin Speckter. Rarely used in practice, but beloved by type enthusiasts.

Standard footnote sequence: * † ‡ § ‖ ¶ — then doubled: ** †† ‡‡ and so on.

The classical Roman alphabet had 23 letters: A B C D E F G H I K L M N O P Q R S T V X — and no J, U, W, Y, or Z in their current roles. The 26-letter alphabet we use today is the result of centuries of additions and distinctions.

I → I and J: The letter I served for both the vowel (as in “it”) and the consonant (as in “yet”). J emerged gradually in the 15th–16th century as a distinct form for the consonant sound. Many early printed books use I where we would write J.

V → U and V: V was used for both the vowel (u) and the consonant (v). The rounded form U was distinguished during the Renaissance. Classical inscriptions like IVLIVS CAESAR render what we’d write as Julius Caesar.

W: Invented for Germanic and Anglo-Saxon languages that had a /w/ sound that Latin lacked. It is literally a doubled V (or U) — hence the name “double-u,” even though its modern form often looks like a double-V.

Y: Borrowed from the Greek letter Upsilon (Υ) to write Greek loanwords in Latin. In English, Y became both vowel (gym, by) and consonant (yes, year).

Z: Also from Greek (Zeta, Ζ). Originally position 7 in the early Latin alphabet; then dropped for centuries because Latin had no use for the /z/ sound; reinstated at the end of the alphabet when Greek loanwords required it — which is why Z comes last despite being an ancient letter.

G: An invention within Latin. Early Latin used C for both the /k/ and /g/ sounds. Around 230 BCE, a Roman schoolteacher named Spurius Carvilius Ruga added a small bar to C to create G, giving it position 7 (the spot Z had vacated).

The Latin alphabet was designed for one language. When other languages adopted it, they each repurposed the available letters to represent their own sounds — often very differently from one another.

G: English /g/ (go) or /dʒ/ (gem). Dutch: the guttural /x/ (like the Scottish loch). Italian: /g/ or /dʒ/ (gelato). Spanish: /x/ before e/i (general), hard /g/ elsewhere.

J: English /dʒ/ (jump). Spanish /x/ (Juan). French /ʒ/ (bonjour). German /j/ (Jahr — a Y-sound). Dutch /j/ as well.

LL: Spanish traditionally /ʎ/ (like Italian gli), though most speakers today say /j/ (like English Y). Welsh: a voiceless lateral fricative /ɬ/ — a sound with no English equivalent, made by placing the tongue for L and blowing sideways.

W: English /w/. German /v/ (Wasser = water). Welsh uses W as a vowel /ʊ/ or /uː/.

H: English /h/. Spanish: fully silent (hola, hora). French: mostly silent. Romanian: pronounced.

C: English /k/ (cat) or /s/ (city). Italian /k/ or /tʃ/ (ciao). Turkish /dʒ/. Polish /ts/.

Why this matters for Hebrew speakers: Hebrew is largely phonemic — each letter represents one consistent sound. The Latin alphabet is historical and full of context-dependent pronunciations. When a Hebrew speaker learns a European language, they often assume the letters are as consistent as Hebrew’s — and are surprised that the same letter can sound completely different depending on what follows it.

Before digital type, measurements had to be precise enough for metal to fit together perfectly in a press. Two separate systems developed — one in Britain and America, one in continental Europe — and they never fully merged.

Point (pt) — the fundamental unit of typographic measurement. In the Anglo-American system: 1 pt = 1⁄72 inch ≈ 0.353 mm. In PostScript and CSS: exactly 1⁄72 inch. Font sizes, leading, and stroke weights are all measured in points.

Pica — 12 points = 1 pica ≈ 4.233 mm. Used to measure column widths, text area dimensions, and margins. There are 6 picas to an inch (approximately).

Didot Point — the continental European point, slightly larger than the Anglo-American point: 1 Didot pt ≈ 0.376 mm. Developed by François-Ambroise Didot in 18th-century France. Still standard in France, Germany, and much of Europe.

Cicero — 12 Didot points = 1 cicero ≈ 4.511 mm. The European equivalent of the pica. Named after a famous edition of Cicero’s writings set in this size in 1467.

Agate — 5.5 points. Used almost exclusively in American newspaper publishing to measure column depth (classified ads were sold by the agate line).

Em — a relative unit equal to the current font size. If the type is set at 12pt, 1 em = 12pt. Originally defined as the width of the capital M (which was typically square). Used for indents, horizontal spacing, and CSS layout.

En — half an em. The en dash (–) is nominally one en wide. Used for spacing and dash lengths.

Thin space — typically 1⁄5 or 1⁄4 of an em. Used in fine typography between a number and its unit, or around certain punctuation in French (« like this »).

Old named sizes: before standardized point sizes, each type size had its own name. A few that survive in common language: Nonpareil (6pt), Brevier (8pt), Bourgeois (9pt), Long Primer (10pt), Pica (12pt, also a size name), English (14pt), Great Primer (18pt). These names were used across print shops because point sizes weren’t yet standardized — the size of a “pica” body varied from one foundry to another.