Orthography is the set of conventions used to write a language. In its narrower sense, it means the accepted spelling system; more broadly, it includes punctuation, capitalization, hyphenation, and word boundaries. The term also names the study of these conventions. Orthography concerns how written forms are organized and used, rather than simply the physical shapes of letters. Its name derives ultimately from Greek orthos, “correct,” and -graphia, “writing.” (oxfordlearnersdictionaries.com)
Orthography, scripts, and writing systems
A writing system provides the means of representing language graphically, while an orthography specifies conventions for applying those means to a particular language. A script can therefore support several orthographies. The Latin alphabet, for example, supplies characters used in English and many other languages, but their spelling rules and sound correspondences differ. Choosing a script does not by itself determine how words should be written. (en.wikipedia.org)
Orthography is not limited to an alphabet. It also applies to syllabic and character-based writing. Japanese combines kanji, historically derived from Chinese characters, with the kana syllabaries. Its conventions govern not only individual written forms but also the distribution of different character types within words and texts. Consequently, orthographic description must accommodate systems in which written units do not correspond simply to individual speech sounds. (en.wikipedia.org)
Components and linguistic representation
A grapheme is a contrastive unit of writing. Orthographic analysis examines the relationships between written units and linguistic units, including phonemes and syllables. A diacritic may distinguish sounds or mark features such as tone, stress, or vowel length. Other conventions determine permissible character combinations, the treatment of borrowed words, and where words may be divided across lines. (scripts.sil.org)
Word boundaries are another central component. Deciding whether an expression is written as one word, separate words, or a hyphenated sequence may require analysis of phonology, morphology, and sometimes syntax. Written boundaries are therefore not merely records of pauses in speech. An orthography must establish usable conventions even where different linguistic criteria suggest different divisions. (mexico.sil.org)
Punctuation and capitalization organize written discourse and distinguish particular categories of words. Their rules vary between languages and writing traditions; capitalization is relevant only where the writing system distinguishes letter cases. These conventions belong to orthography in its broad sense, although specialized spelling manuals may treat them separately. (en.wikipedia.org)
Sound, word structure, and orthographic depth
Orthographies differ in the linguistic level they represent. A phonemic approach emphasizes distinctions between phonemes rather than every detail of pronunciation. A morphophonemic approach may preserve relationships among forms of a morpheme despite changes in their spoken realization. Orthographic design thus involves choices about whether to represent underlying patterns or forms closer to actual pronunciation. These approaches need not be mutually exclusive within one system. (scripts.sil.org)
Orthographic depth describes the complexity and predictability of spelling–sound relationships, especially in alphabetic systems. Relatively shallow, or transparent, orthographies have more consistent correspondences; deeper orthographies require greater use of context and word-specific knowledge. Finnish is commonly contrasted with English in this respect. Depth is not simply a count of irregular words: multiletter units, alternative sound values, and contextual rules can contribute different dimensions of complexity. (frontiersin.org)
Reading and spelling also involve opposite directions of correspondence: deriving pronunciation from writing and selecting a written form from pronunciation. A system may be more predictable in one direction than the other. Research on literacy acquisition has associated orthographic depth with differences in early word and nonword reading. A 2003 comparison of English and twelve other European orthographies reported slower foundation-level reading development in deeper systems, particularly English. These findings concern specific reading tasks rather than a general ranking of languages. (frontiersin.org)
Standardization and variation
Orthographic standardization establishes relatively stable written forms across writers and publications. Its historical development can be gradual: early modern English exhibited substantial spelling variation before the range of alternatives narrowed. Printing contributed to this process, but research treats its role as something to investigate through particular printers, texts, and graphemic developments, rather than as a single event that instantly fixed spelling. (eprints.lancs.ac.uk)
Dialect differences complicate standardization because speakers may pronounce corresponding words differently. A shared orthography can select forms readable across several dialects rather than reproduce one variety in every detail. Social acceptance is also important: technically consistent conventions may fail to gain widespread use if intended readers and writers reject them. Orthographic development therefore includes linguistic analysis, community participation, and testing of readability and writability. (scholars.sil.org)
Digital representation
Digital orthography requires distinctions between linguistic conventions, character encoding, and visual rendering. Unicode provides standardized character representations, but an orthographic description must additionally identify the characters used, their combinations, text direction, and input requirements. Fonts and keyboards are necessary implementation resources; they do not themselves establish spelling rules. (scripts.sil.org)
The same written element can sometimes have more than one equivalent digital representation. For example, an accented letter may be encoded as a precomposed character or as a base letter followed by a combining mark. Unicode normalization makes such equivalent sequences consistent for processing. This is distinct from orthographic standardization: normalization addresses encoded representations, not decisions about which spelling a language community accepts. (unicode.org)