Letter case

Capitalisation is the writing of a word with its first letter in uppercase and the remaining letters in lowercase. Capitalisation rules vary by language and are often quite complex, but in most modern languages that have capitalisation, the first word of every sentence is capitalised, as are all proper nouns. [8]

Capitalisation in English, in terms of the general orthographic rules independent of context (e.g. title vs. heading vs. text), is universally standardised for formal writing. Capital letters are used as the first letter of a sentence, a proper noun, or a proper adjective. The names of the days of the week and the names of the months are also capitalised, as are the first-person pronoun "I"[9] and the vocative particle "O". There are a few pairs of words of different meanings whose only difference is capitalisation of the first letter. Honorifics and personal titles showing rank or prestige are capitalised when used together with the name of the person (for example, "Mr. Smith", "Bishop O'Brien", "Professor Moore") or as a direct address, but normally not when used alone and in a more general sense.[10][11] It can also be seen as customary to capitalise any word – in some contexts even a pronoun[12] – referring to the deity of a monotheistic religion.

Other words normally start with a lower-case letter. There are, however, situations where further capitalisation may be used to give added emphasis, for example in headings and publication titles (see below). In some traditional forms of poetry, capitalisation has conventionally been used as a marker to indicate the beginning of a line of verse independent of any grammatical feature. In political writing, parody and satire, the unexpected emphasis afforded by otherwise ill-advised capitalisation is often used to great stylistic effect, such as in the case of George Orwell's Big Brother.

Other languages vary in their use of capitals. For example, in German all nouns are capitalised (this was previously common in English as well, mainly in the 17th and 18th centuries), while in Romance and most other European languages the names of the days of the week, the names of the months, and adjectives of nationality, religion, and so on normally begin with a lower-case letter.[13] On the other hand, in some languages it is customary to capitalise formal polite pronouns, for example De, Dem (Danish), Sie, Ihnen (German), and Vd or Ud (short for usted in Spanish).

Informal communication, such as texting, instant messaging or a handwritten sticky note, may not bother to follow the conventions concerning capitalisation, but that is because its users usually do not expect it to be formal.[9]

• The German letter "ß" formerly existed only in lower case. The orthographical capitalisation does not concern "ß", which generally does not occur at the beginning of a word, and in the all-caps style it has traditionally been replaced by the digraph "SS". Since June 2017, however, capital ẞ is accepted as an alternative in the all-caps style.[14]
• The Greek upper-case letter "Σ" has two different lower-case forms: "ς" in word-final position and "σ" elsewhere. In a similar manner, the Latin upper-case letter "S" used to have two different lower-case forms: "s" in word-final position and " ſ " elsewhere. The latter form, called the long s, fell out of general use before the middle of the 19th century, except for the countries that continued to use blackletter typefaces such as Fraktur. When blackletter type fell out of general use in the mid-20th century, even those countries dropped the long s.
• The treatment of the Greek iota subscript with upper-case letters is complicated.
• Unlike most languages that use Latin-script and link the dotless upper-case "I" with the dotted lower-case "i", Turkish as well as some forms of Azeri have both a dotted and dotless I, each in both upper and lower case. Each of the two pairs ("İ/i" and "I/ı") represents a distinctive phoneme.
• In some languages, specific digraphs may be regarded as single letters, and in Dutch, the digraph "IJ/ij" is even capitalised with both components written in uppercase (for example, "IJsland" rather than "Ijsland").[15] In other languages, such as Welsh and Hungarian, various digraphs are regarded as single letters for collation purposes, but the second component of the digraph will still be written in lower case even if the first component is capitalised. Similarly, in South Slavic languages whose orthography is coordinated between the Cyrillic and Latin scripts, the Latin digraphs "Lj/lj", "Nj/nj" and "Dž/dž" are each regarded as a single letter (like their Cyrillic equivalents "Љ/љ", "Њ/њ" and "Џ/џ", respectively), but only in all-caps style should both components be in upper case (e.g. Ljiljan–LJILJAN, Njonja–NJONJA, Džidža–DŽIDŽA). Unicode designates a single character for each case variant (i.e., upper case, title case and lower case) of the three digraphs.[16]
• Some English surnames such as fforbes are traditionally spelt with a digraph instead of a capital letter (at least for ff). This indicates a long and prestigious family tradition.
• In the Hawaiian orthography, the ʻokina is a phonemic symbol that visually resembles a left single quotation mark. Representing the glottal stop, the ʻokina can be characterised as either a letter[17] or a diacritic.[18] As a unicase letter, the ʻokina is unaffected by capitalisation; it is the following letter that is capitalised instead. According to the Unicode standard, the ʻokina is formally encoded as U+02BB ʻ MODIFIER LETTER TURNED COMMA,[19] but it is not uncommon to substitute this with a similar punctuation character, such as the left single quotation mark or an apostrophe.[20]

Similar orthographic and graphostylistic conventions are used for emphasis or following language-specific or other rules, including:

• Font effects such as italic type or oblique type, boldface, and choice of serif vs. sans-serif.
• Typographical conventions in mathematical formulae include the use of Greek letters and the use of Latin letters with special formatting such as blackboard bold and blackletter.
• Some letters of the Arabic and Hebrew alphabets and some jamo of the Korean hangul have different forms depending on placement within a word, but these rules are strict and the different forms cannot be used for emphasis.
  • In the Arabic and Arabic-based alphabets, letters in a word are connected, except for several that cannot connect to the following letter. Letters may have distinct forms depending on whether they are initial (connected only to the following letter), medial (connected to both neighboring letters), final (connected only to the preceding letter), or isolated (connected to neither a preceding nor a following letter).
  • In the Hebrew alphabet, five letters have a distinct form (see Final form) that is used when they are word-final.
• In Georgian, some authors use isolated letters from the ancient Asomtavruli alphabet within a text otherwise written in the modern Mkhedruli in a fashion that is reminiscent of the usage of upper-case letters in the Latin, Greek, and Cyrillic alphabets.
• In the Japanese writing system, an author has the option of switching between kanji, hiragana, katakana, and rōmaji. In particular, every hiragana character has an equivalent katakana character, and vice versa. Romanised Japanese sometimes uses lowercase letters to represent words that would be written in hiragana, and uppercase letters to represent words that would be written in katakana. Some kana characters are written in smaller type when they modify or combine with the preceding sign (yōon) or the following sign (sokuon).

In English-language publications, various conventions are used for the capitalisation of words in publication titles and headlines, including chapter and section headings. The rules differ substantially between individual house styles.

The convention followed by many British publishers (including scientific publishers like Nature and New Scientist, magazines like The Economist, and newspapers like The Guardian and The Times) and many U.S. newspapers is sentence-style capitalisation in headlines, i.e. capitalisation follows the same rules that apply for sentences. This convention is usually called sentence case. It may also be applied to publication titles, especially in bibliographic references and library catalogues. An example of a global publisher whose English-language house style prescribes sentence-case titles and headings is the International Organization for Standardization (ISO).

For publication titles it is, however, a common typographic practice among both British[24] and U.S. publishers to capitalise significant words (and in the United States, this is often applied to headings, too). This family of typographic conventions is usually called title case. For example, R. M. Ritter's Oxford Manual of Style (2002) suggests capitalising "the first word and all nouns, pronouns, adjectives, verbs and adverbs, but generally not articles, conjunctions and short prepositions".[25] This is an old form of emphasis, similar to the more modern practice of using a larger or boldface font for titles. The rules which prescribe which words to capitalise are not based on any grammatically inherent correct–incorrect distinction and are not universally standardised; they differ between style guides, although most style guides tend to follow a few strong conventions, as follows:

• Most styles capitalise all words except for short closed-class words (certain parts of speech, namely, articles, prepositions, and conjunctions); but the first word (always) and last word (in many styles) are also capitalised, regardless of their part of speech. Many styles capitalise longer prepositions such as "between" and "throughout", but not shorter ones such as "for" and "with".[26] Typically, a preposition is considered short if it has up to three or four letters.
• A few styles capitalise all words in title case (the so-called start case), which has the advantage of being easy to implement and hard to get "wrong" (that is, "not edited to style"). Because of this rule's simplicity, software case-folding routines can handle 95% or more of the editing, especially if they are programmed for desired exceptions (such as "FBI" rather than "Fbi").
• As for whether hyphenated words are capitalised not only at the beginning but also after the hyphen, there is no universal standard; variation occurs in the wild and among house styles (e.g., "The Letter-Case Rule in My Book"; "Short-term Follow-up Care for Burns"). Traditional copyediting makes a distinction between temporary compounds (such as many nonce [novel instance] compound modifiers), in which every part of the hyphenated word is capitalised (e.g. "How This Particular Author Chose to Style His Autumn-Apple-Picking Heading"), and permanent compounds, which are terms that, although compound and hyphenated, are so well established that dictionaries enter them as headwords (e.g., "Short-term Follow-up Care for Burns").

Title case is widely used in many English-language publications, especially in the United States. However, its conventions are sometimes not followed strictly – especially in informal writing.

In creative typography, such as music record covers and other artistic material, all styles are commonly encountered, including all-lowercase letters and special case styles, such as studly caps (see below). For example, in the wordmarks of video games it is not uncommon to use stylised upper-case letters at the beginning and end of a title, with the intermediate letters in small caps or lower case (e.g., ArcaniA, ArmA, and DmC).

Single-word proper nouns are capitalised in formal written English, unless the name is intentionally stylised to break this rule (such as the first or last name of danah boyd).

Multi-word proper nouns include names of organisations, publications, and people. Often the rules for "title case" (described in the previous section) are applied to these names, so that non-initial articles, conjunctions, and short prepositions are lowercase, and all other words are uppercase. For example, the short preposition "of" and the article "the" are lowercase in "Steering Committee of the Finance Department". Usually only capitalised words are used to form an acronym variant of the name, though there is some variation in this.

With personal names, this practice can vary (sometimes all words are capitalised, regardless of length or function), but is not limited to English names. Examples include the English names Tamar of Georgia and Catherine the Great, "van" and "der" in Dutch names, "von" and "zu" in German, "de", "los", and "y" in Spanish names, "de" or "d'" in French names, and "ibn" in Arabic names.

Some surname prefixes also affect the capitalisation of the following internal letter or word, for example "Mac" in Celtic names and "Al" in Arabic names.

Of the seven SI base-unit symbols, "A" (ampere for electric current) and "K" (kelvin for temperature), both named after people, are always written in upper case, whereas "s" (second for time), "m" (metre for length), "kg" (kilogram for mass), "cd" (candela for luminous intensity), and "mol" (mole for amount of substance) are written in lower case.

In the International System of Units (SI), a letter usually has different meanings in upper and lower case when used as a unit symbol. Generally, unit symbols are written in lower case, but if the name of the unit is derived from a proper noun, the first letter of the symbol is capitalised. Nevertheless, the name of the unit, if spelled out, is always considered a common noun and written accordingly in lower case.[27] For example:

• 1 s (one second) when used for the base unit of time.
• 1 S (one siemens) when used for the unit of electric conductance and admittance (named after Werner von Siemens).
• 1 Sv (one sievert), used for the unit of ionising radiation dose (named after Rolf Maximilian Sievert).

For the purpose of clarity, the symbol for litre can optionally be written in upper case even though the name is not derived from a proper noun.[27] For example, "one litre" may be written as:

• 1 l, the original form, for typefaces in which "digit one" ⟨1⟩, "lower-case ell" ⟨l⟩, and "upper-case i" ⟨I⟩ look different.
• 1 L, an alternative form, for typefaces in which these characters are difficult to distinguish, or the typeface the reader will be using is unknown. A "script l" in various typefaces (e.g.: 1 l) has traditionally been used in some countries to prevent confusion; however, the separate Unicode character which represents this, U+2113 ℓ SCRIPT SMALL L, is deprecated by the SI.[28] Another solution sometimes seen in Web typography is to use a serif font for "lower-case ell" in otherwise sans-serif material (1 l).

The letter case of a prefix symbol is determined independently of the unit symbol to which it is attached. Lower case is used for all submultiple prefix symbols and the small multiple prefix symbols up to "k" (for kilo, meaning 10³ = 1000 multiplier), whereas upper case is used for larger multipliers:[27]

• 1 ms, millisecond, a small measure of time ("m" for milli, meaning 10⁻³ = 1/1000 multiplier).
• 1 Ms, megasecond, a large measure of time ("M" for mega, meaning 10⁶ = 1 000 000 multiplier).
• 1 mS, millisiemens, a small measure of electric conductance.
• 1 MS, megasiemens, a large measure of electric conductance.
• 1 mm, millimetre, a small measure of length.
• 1 Mm, megametre, a large measure of length.

Some case styles are not used in standard English, but are common in computer programming, product branding, or other specialised fields.

The usage derives from how programming languages are parsed, programmatically. They generally separate their syntactic tokens by simple whitespace, including space characters, tabs, and newlines. When the tokens, such as function and variable names start to multiply in complex software development, and there is still a need to keep the source code human-readable, Naming conventions make this possible. So for example, a function dealing with matrix multiplication might formally be called:

• • SGEMM(*), with the asterisk standing in for an equally inscrutable list of 13 parameters (in BLAS),
  • MultiplyMatrixByMatrix(Matrix x, Matrix y), in some hypothetical higher level manifestly typed language, broadly following the syntax of C++ or Java,
  • multiply-matrix-by-matrix(x, y) in something derived from LISP, or perhaps
  • (multiply (x y)) in the CLOS, or some newer derivative language supporting type inference and multiple dispatch.

In each case the capitalisation or lack thereof supports a different function. In the first, FORTRAN compatibility requires case-insensitive naming and short function names. The second supports easily discernible function and argument names and types, within the context of an imperative, strongly typed language. The third supports the macro facilities of LISP, and its tendency to view programs and data minimalistically, and as interchangeable. The fourth idiom needs much less syntactic sugar overall, because much of the semantics are implied, but because of its brevity and so lack of the need for capitalization or multipart words at all, might also make the code too abstract and overloaded for the common programmer to understand.

Understandably then, such coding conventions are highly subjective, and can lead to rather opinionated debate, such as in the case of editor wars, or those about indent style. Capitalisation is no exception.

Camel case: "theQuickBrownFoxJumpsOverTheLazyDog" or "TheQuickBrownFoxJumpsOverTheLazyDog"
Spaces and punctuation are removed and the first letter of each word is capitalised. If this includes the first letter of the first word (CamelCase, "PowerPoint", "TheQuick...", etc.), the case is sometimes called upper camel case (or, illustratively, CamelCase), Pascal case in reference to the Pascal programming language[29] or bumpy case.

When the first letter of the first word is lowercase ("iPod", "eBay", "theQuickBrownFox..."), the case is usually known as lower camel case or dromedary case (illustratively: dromedaryCase). This format has become popular in the branding of information technology products and services, with an initial "i" meaning "Internet" or "intelligent", as in iPod, or an initial "e" meaning "electronic", as in email (electronic mail) or e-commerce (electronic commerce).

Snake case: "the_quick_brown_fox_jumps_over_the_lazy_dog"
Punctuation is removed and spaces are replaced by single underscores. Normally the letters share the same case (e.g. "UPPER_CASE_EMBEDDED_UNDERSCORE" or "lower_case_embedded_underscore") but the case can be mixed, as in OCaml modules.[30] The style may also be called pothole case, especially in Python programming, in which this convention is often used for naming variables. Illustratively, it may be rendered snake_case, pothole_case, etc. When all-upper-case, it may be referred to as screaming snake case (or SCREAMING_SNAKE_CASE) or hazard case.[31]

Kebab case: "the-quick-brown-fox-jumps-over-the-lazy-dog"
Similar to snake case, above, except hyphens rather than underscores are used to replace spaces. It is also known as spinal case, param case, Lisp case in reference to the Lisp programming language, or dash case (or illustratively as kebab-case). If every word is capitalised, the style is known as train case (TRAIN-CASE).

In CSS, all property names and most keyword values are primarily formatted in kebab case.

Studly caps: e.g. "tHeqUicKBrOWnFoXJUmpsoVeRThElAzydOG"
Mixed case with no semantic or syntactic significance to the use of the capitals. Sometimes only vowels are upper case, at other times upper and lower case are alternated, but often it is simply random. The name comes from the sarcastic or ironic implication that it was used in an attempt by the writer to convey their own coolness. It is also used to mock the violation of standard English case conventions by marketers in the naming of computer software packages, even when there is no technical requirement to do so – e.g., Sun Microsystems' naming of a windowing system NeWS. Illustrative naming of the style is, naturally, random: stUdlY cAps, StUdLy CaPs, etc.

Unicode defines case folding through the three case-mapping properties of each character: upper case, lower case, and title case (in this context, "title case" relates to ligatures and digraphs encoded as mixed-case single characters, in which the first component is in upper case and the second component in lower case[32]). These properties relate all characters in scripts with differing cases to the other case variants of the character.

As briefly discussed in Unicode Technical Note #26,[33] "In terms of implementation issues, any attempt at a unification of Latin, Greek, and Cyrillic would wreak havoc [and] make casing operations an unholy mess, in effect making all casing operations context sensitive […]". In other words, while the shapes of letters like A, B, E, H, K, M, O, P, T, X, Y and so on are shared between the Latin, Greek, and Cyrillic alphabets (and small differences in their canonical forms may be considered to be of a merely typographical nature), it would still be problematic for a multilingual character set or a font to provide only a single code point for, say, uppercase letter B, as this would make it quite difficult for a wordprocessor to change that single uppercase letter to one of the three different choices for the lower-case letter, the Latin b (U+0062), Greek β (U+03B2) or Cyrillic в (U+0432). Therefore, the corresponding Latin, Greek and Cyrillic upper-case letters (U+0042, U+0392 and U+0412, respectively) are also encoded as separate characters, despite their appearance being basically identical. Without letter case, a "unified European alphabet" – such as ABБCГDΔΕЄЗFΦGHIИJ...Z, with an appropriate subset for each language – is feasible; but considering letter case, it becomes very clear that these alphabets are rather distinct sets of symbols.

Most modern word processors provide automated case conversion with a simple click or keystroke. For example, in Microsoft Office Word, there is a dialog box for toggling the selected text through UPPERCASE, then lowercase, then Title Case (actually start caps; exception words must be lowercased individually). The keystroke ⇧ Shift+F3 does the same thing.

In some forms of BASIC there are two methods for case conversion:

 UpperA$ = UCASE$("a")
 LowerA$ = LCASE$("A")

C and C++, as well as any C-like language that conforms to its standard library, provide these functions in the file ctype.h:

 char upperA = toupper('a');
 char lowerA = tolower('A');

Case conversion is different with different character sets. In ASCII or EBCDIC, case can be converted in the following way, in C:

 #define toupper(c) (islower(c) ? (c) – 'a' + 'A' : (c))
 #define tolower(c) (isupper(c) ? (c) – 'A' + 'a' : (c))

This only works because the letters of upper and lower cases are spaced out equally. In ASCII they are consecutive, whereas with EBCDIC they are not; nonetheless the upper-case letters are arranged in the same pattern and with the same gaps as are the lower-case letters, so the technique still works.

Some computer programming languages offer facilities for converting text to a form in which all words are capitalised. Visual Basic calls this "proper case"; Python calls it "title case". This differs from usual title casing conventions, such as the English convention in which minor words are not capitalised.

The individual type blocks used in hand typesetting are stored in shallow wooden or metal drawers known as "type cases". Each is subdivided into a number of compartments ("boxes") for the storage of different individual letters.

The Oxford Universal Dictionary on Historical Advanced Proportional Principles (reprinted 1952) indicates that case in this sense (referring to the box or frame used by a compositor in the printing trade) was first used in English in 1588. Originally one large case was used for each typeface, then "divided cases", pairs of cases for majuscules and minuscules, were introduced in the region of today's Belgium by 1563, England by 1588, and France before 1723.

The terms upper and lower case originate from this division. By convention, when the two cases were taken out of the storage rack and placed on a rack on the compositor's desk, the case containing the capitals and small capitals stood at a steeper angle at the back of the desk, with the case for the small letters, punctuation, and spaces being more easily reached at a shallower angle below it to the front of the desk, hence upper and lower case.[38]

Though pairs of cases were used in English-speaking countries and many European countries in the seventeenth century, in Germany and Scandinavia the single case continued in use.[38]

Various patterns of cases are available, often with the compartments for lower-case letters varying in size according to the frequency of use of letters, so that the commonest letters are grouped together in larger boxes at the centre of the case.[38] The compositor takes the letter blocks from the compartments and places them in a composing stick, working from left to right and placing the letters upside down with the nick to the top, then sets the assembled type in a galley.[38]