11.8 Historical linguistics

Origins of historical linguistics

Historical linguistics explores how languages evolve over time. By studying these changes, researchers gain insight into human migration patterns, cultural development, and the connections between societies separated by centuries. The field sits at the intersection of linguistics, anthropology, and history.

Three core methods drive this work: the comparative method, internal reconstruction, and the analysis of sound change principles. Together, these tools let linguists trace language families, reconstruct earlier forms of languages, and understand how languages have shaped each other across time.

Comparative method

The comparative method works by systematically comparing related languages to reconstruct their common ancestor. Linguists look for cognates, words in different languages that share a common origin, and identify regular sound correspondences between them.

For example, comparing Latin pater, English father, and Sanskrit pitar reveals a consistent pattern in how certain consonants shifted across these languages. These patterns point back to a shared ancestor: Proto-Indo-European. By cataloging enough of these regular correspondences, linguists can reconstruct vocabulary, sounds, and even some grammar of languages that were never written down.

Internal reconstruction

Sometimes there aren't enough related languages available for comparison. In those cases, linguists turn inward, analyzing patterns and irregularities within a single language to figure out what earlier stages might have looked like.

Irregularities in modern word forms often preserve traces of older rules. If a language has an unexpected alternation in its verb forms, for instance, that inconsistency may reflect a sound change that happened centuries ago. Internal reconstruction can reveal past sound changes, shifts in word structure, and developments in meaning, all from evidence within one language.

Sound change principles

A foundational idea in historical linguistics is the Neogrammarian hypothesis: sound changes are regular and apply without exception across all relevant words in a language. This regularity is what makes reconstruction possible.

Two classic examples:

• Grimm's Law describes a systematic set of consonant shifts that occurred as Proto-Indo-European evolved into Proto-Germanic. For instance, the PIE p sound shifted to f in Germanic languages, which is why Latin piscis corresponds to English fish.
• Verner's Law explains apparent exceptions to Grimm's Law by showing that stress placement affected which consonants shifted and which didn't.

These principles give linguists predictive power: if you know the rules, you can often predict what a word should look like in a related language.

Language families and classification

Linguists organize the world's roughly 7,000 languages into families based on shared origins. This classification reveals historical relationships between languages and, by extension, between the peoples who spoke them.

Indo-European language family

The Indo-European family is the largest and most studied, encompassing languages spoken by nearly half the world's population. It includes English, Spanish, Hindi, Russian, and many others.

Proto-Indo-European is believed to have been spoken around 4000–6000 BCE, most likely in the Pontic-Caspian steppe region north of the Black Sea. From that single ancestor, the family branched into about ten major groups, including Germanic, Romance, Slavic, and Indo-Iranian. Reconstructing Proto-Indo-European has been one of historical linguistics' greatest achievements, providing a window into a culture that left no written records.

Other major language families

• Sino-Tibetan includes the Chinese languages (Mandarin, Cantonese, etc.) and Tibeto-Burman languages like Burmese and Tibetan. It's the second-largest family by number of speakers.
• Afroasiatic encompasses Semitic languages (Arabic, Hebrew), Berber languages, and ancient Egyptian.
• Niger-Congo is the largest African language family, containing the vast Bantu subfamily (Swahili, Zulu, and hundreds more).
• Austronesian stretches from Madagascar off the African coast all the way to Easter Island in the Pacific, including Malay, Tagalog, and the Polynesian languages.

Methods of language classification

Not all classification works the same way:

• Genetic classification groups languages by common ancestry, using shared innovations to build family trees.
• Typological classification categorizes languages by structural features like word order (subject-verb-object vs. subject-object-verb) or how they build words (isolating vs. agglutinating).
• Areal classification accounts for similarities caused by geographic proximity and prolonged contact rather than shared ancestry.
• Lexicostatistics uses quantitative comparison of shared vocabulary to estimate how closely related two languages are.

Phonological change

Phonological changes are alterations in a language's sound system over time. They can affect individual sounds, combinations of sounds, or entire patterns of pronunciation. These changes are a major reason why related languages sound so different from each other today.

Regular sound changes

Regular sound changes apply systematically to all words containing a particular sound in a given environment. Their regularity is what makes the comparative method work.

• The Great Vowel Shift in English (roughly 1400–1700 CE) dramatically altered the pronunciation of all long vowels. The word bite, for example, was once pronounced more like "beet." This shift is a major reason English spelling seems so disconnected from pronunciation today.
• Grimm's Law (mentioned above) describes consonant shifts from Proto-Indo-European to Proto-Germanic, affecting entire classes of sounds in a predictable way.

Because these changes are regular, linguists can predict corresponding word forms across related languages.

Sporadic sound changes

Unlike regular changes, sporadic changes affect only individual words or small groups of words. They're often driven by specific circumstances rather than system-wide rules.

• Metathesis is the reordering of sounds within a word. Old English thridda became modern third, with the r and i swapping positions.
• Dissimilation occurs when two similar sounds in a word become less alike. Latin peregrinus eventually became English pilgrim, with the first r changing to l.

Phonological processes

Several recurring processes shape how sounds change:

• Assimilation makes neighboring sounds more similar to each other (the n in Latin in- becomes m before p in impossible).
• Lenition weakens consonants, often between vowels. Latin vita ("life") became Spanish vida, with the hard t softening to d.
• Epenthesis inserts a sound into a word. Old English þunor gained a d to become thunder.
• Apocope is the loss of sounds at the end of a word. The final -e in words like name was once pronounced in Middle English but has since gone silent.

Morphological change

Morphological changes affect how words are structured and formed. Over time, languages can gain or lose inflectional endings, develop new ways of building words, and reorganize entire grammatical systems.

Analogy and leveling

Analogy is a powerful force in language change. Speakers extend familiar patterns to places where they didn't originally apply.

A clear example: Old English used holp as the past tense of help. Over time, speakers regularized it to helped, following the dominant pattern of adding -ed. This process, called leveling, reduces irregularity within a language. The logic works like a proportion: if walk : walked :: help : X, then X = helped.

Grammaticalization

Grammaticalization is the process by which ordinary content words gradually become grammatical function words. This typically involves both a loss of specific meaning (called semantic bleaching) and a reduction in pronunciation.

• English going to started as a verb of motion but has become a future tense marker (gonna).
• The Latin demonstrative pronoun ille ("that") evolved into the French definite article le ("the").

Inflectional vs. derivational changes

• Inflectional changes affect grammatical markers like case endings and tense markers. Old English had a complex case system for nouns (nominative, accusative, genitive, dative), but Modern English has largely lost these distinctions, relying on word order instead.
• Derivational changes affect how new words are formed. Languages develop new prefixes and suffixes over time, and existing ones may become more or less productive. The suffix -ize, for instance, has become increasingly productive in Modern English (prioritize, incentivize).

Semantic change

Words don't just change in sound and form; their meanings shift too. Semantic change reflects broader cultural, technological, and social developments, and tracking it helps us understand how societies have evolved.

Types of semantic change

• Broadening expands a word's meaning. Old English dogga referred to one specific breed but eventually came to mean any dog.
• Narrowing restricts a word's meaning. Meat once meant food in general (as preserved in the phrase "meat and drink") but narrowed to refer specifically to animal flesh.
• Amelioration improves a word's connotation. Knight originally meant "servant" or "boy" but rose in status to mean a noble warrior.
• Pejoration worsens a word's connotation. Silly once meant "blessed" or "innocent" in Old English, then gradually shifted to mean "foolish."

Mechanisms of semantic change

• Metaphorical extension applies a concrete meaning to an abstract concept. You can grasp an idea the same way you grasp an object.
• Metonymy substitutes one concept for a closely associated one. "The White House announced..." uses the building to stand for the U.S. government.
• Synecdoche uses a part to represent the whole. Calling a car your "wheels" is a classic example.
• Euphemism replaces uncomfortable terms with softer alternatives, which can then take on the negative connotation themselves over time.

Cultural influences on meaning

Technology, social change, and historical events all reshape word meanings:

• Mouse gained a new meaning with the invention of the computer peripheral.
• Gay shifted from meaning "happy" or "carefree" to its current primary meaning of "homosexual" over the course of the 20th century.
• Holocaust existed as a general term for destruction by fire before gaining its specific association with the genocide of World War II.
• Cross-cultural contact introduces loanwords that carry new meanings into a language, like algebra from Arabic or tsunami from Japanese.

Syntactic change

Syntactic changes alter how sentences are structured and how grammatical relationships are expressed. These shifts can fundamentally reshape how a language works.

Word order shifts

Languages can shift between different basic word orders over time. The main types are SOV (subject-object-verb), SVO (subject-verb-object), and VSO (verb-subject-object).

Old English had relatively flexible word order because its case system made grammatical roles clear regardless of position. As those case endings eroded, English became more dependent on a fixed SVO order. Similarly, Latin's SOV tendency gave way to SVO in the Romance languages. Word order changes and case system changes tend to go hand in hand.

Grammatical category changes

Words can shift from one grammatical category to another over time:

• Back started as a noun (a body part) and developed into a preposition ("in back of") and an adverb ("go back").
• Old English habban was a full verb meaning "to have/possess" but evolved into the auxiliary have used to form perfect tenses ("I have eaten").
• Old English sum, meaning "a certain one," became the modern determiner some.

These shifts frequently involve grammaticalization.

Syntactic reanalysis

Reanalysis happens when listeners interpret a sentence's structure differently from how the speaker intended it. Over time, the new interpretation becomes the standard.

• Old English expressed "it pleases me" as it likes me, with me as the object. Speakers eventually reanalyzed this as "I like it," turning the experiencer into the subject.
• The phrase a napron (from Old French naperon) was reanalyzed as an apron because listeners heard the word boundary differently.

Reanalysis can create entirely new grammatical constructions and even new words.

Lexical change

A language's vocabulary is constantly in flux. New words enter, old words fade, and borrowed words get adapted. These changes reflect what's happening in the culture that uses the language.

Borrowing and loanwords

Languages adopt words from other languages whenever cultures come into contact. English is a particularly heavy borrower, with major contributions from French (after the Norman Conquest: justice, government, beef), Latin and Greek (scientific and academic vocabulary), and dozens of other languages.

Loanwords often fill gaps for new concepts, technologies, or cultural items: sushi from Japanese, yoga from Sanskrit, algebra from Arabic. Once borrowed, words typically get adapted to fit the borrowing language's pronunciation and grammar.

Neologisms and coinages

New words are constantly being created:

• Technology drives many coinages: internet, blog, podcast, selfie.
• Scientific terms are often built from Greek or Latin roots: photosynthesis (Greek photos "light" + synthesis "putting together"), neuroscience.
• Blends combine parts of existing words: smog (smoke + fog), brunch (breakfast + lunch).

Obsolescence and archaisms

Words also fall out of use. Thou and wherefore survive mainly in Shakespeare and religious texts. Floppy disk is becoming obsolete as the technology disappears. Studying these archaisms is valuable because they help us read and understand historical texts, and they preserve traces of earlier stages of the language.

Writing systems and historical linguistics

Writing systems are among the most important sources of evidence for historical linguists. Written records preserve snapshots of languages at specific moments in time, making it possible to track changes across centuries.

Development of writing systems

Writing systems evolved from concrete to increasingly abstract representations:

1. Pictographic systems used pictures to represent objects and ideas directly.
2. These evolved into logographic systems (like Chinese characters) where symbols represent words or morphemes.
3. Syllabic systems (like Japanese kana) use symbols for syllables.
4. Alphabetic systems represent individual sounds. The first true alphabet was the Phoenician script (around 1050 BCE), which became the ancestor of Greek, Latin, Arabic, and Hebrew scripts.

Key early writing systems include cuneiform (developed in Mesopotamia around 3200 BCE for Sumerian) and Egyptian hieroglyphs (emerging around 3000 BCE, combining logographic and phonetic elements).

Decipherment of ancient scripts

Deciphering lost scripts has unlocked entire civilizations:

• The Rosetta Stone, with its parallel Greek, Demotic, and hieroglyphic texts, enabled Jean-François Champollion to crack Egyptian hieroglyphs in 1822.
• Linear B was deciphered by Michael Ventris in 1952, revealing it as an early form of Greek and pushing knowledge of Greek back several centuries.
• Mayan script decipherment began in the mid-20th century and continues today, progressively revealing the history and culture of Mesoamerican civilizations.

Orthographic changes over time

Spelling doesn't stay fixed either. The Great Vowel Shift left English with spellings that reflected older pronunciations, which is why English spelling often seems illogical today. Some languages have undertaken deliberate spelling reforms: Germany's 1996 orthography reform standardized various spelling rules, and France's Académie française has long regulated official spelling and vocabulary.

Historical sociolinguistics

Historical sociolinguistics examines how social factors drive language change. Languages don't evolve in a vacuum; they change because of who speaks them, who they're in contact with, and what social pressures surround them.

Language contact and change

When speakers of different languages interact regularly, their languages influence each other:

• Pidgins are simplified contact languages that develop when groups without a shared language need to communicate (often in trade contexts). When a pidgin becomes a community's first language, it develops into a creole with full grammatical complexity.
• Substrate influence occurs when a community shifts to a new language but carries over features from their original one. Irish English, for example, preserves some syntactic patterns from Irish Gaelic.
• A sprachbund (linguistic area) forms when unrelated languages in close geographic proximity develop shared features through prolonged contact. The Balkan sprachbund is a classic example, where Romanian, Bulgarian, Albanian, and Greek share grammatical features despite belonging to different branches of Indo-European.

Prestige and language evolution

Social prestige plays a major role in which linguistic forms spread and which ones fade:

• After the Norman Conquest of 1066, French became the prestige language in England. This led to massive French vocabulary entering English, especially in domains like law, government, and cuisine.
• Hypercorrection occurs when speakers overapply prestige forms, sometimes producing forms that don't actually exist in the prestige variety.
• Language attitudes can speed up or slow down the adoption of new linguistic features.

Language policy and planning

Governments and institutions actively shape language through policy:

• Standardization creates uniform written and spoken norms, often based on a particular dialect.
• Language revival efforts attempt to reverse language decline. The revival of Hebrew as a spoken language in the 20th century is one of the most successful examples.
• Purism movements try to remove foreign influences from a language, as with efforts in Icelandic to coin native terms rather than borrow international ones.

Methods in historical linguistics

Historical linguists rely on several complementary methods to reconstruct the past. Each has strengths and limitations, and the best results come from combining approaches.

Comparative reconstruction

This is the primary method. The process works roughly like this:

1. Identify cognates (words with shared origins) across related languages.
2. Catalog the regular sound correspondences between those cognates.
3. Use those correspondences to reconstruct proto-forms, the hypothetical ancestor words.
4. Build up a picture of the proto-language's sound system, vocabulary, and grammar.

The result can be a partial dictionary and grammar of a language that was never written down.

Internal reconstruction

Internal reconstruction complements the comparative method by working within a single language. Linguists examine alternations in morphological paradigms (like irregular verb forms) to deduce what earlier, more regular forms might have looked like. This method is especially valuable for language isolates, languages with no known relatives, where comparison isn't possible.

Glottochronology and lexicostatistics

These quantitative methods attempt to estimate when related languages diverged:

• Lexicostatistics compares the percentage of shared cognates in basic vocabulary (words for body parts, numbers, kinship terms) between two languages.
• Glottochronology goes further, assuming that basic vocabulary changes at a roughly constant rate (about 14% per millennium) and using that rate to calculate divergence dates.

Both methods are controversial. The assumption of a constant rate of change doesn't always hold, and it can be difficult to distinguish inherited words from borrowings.

Applications of historical linguistics

Historical linguistics contributes far beyond the study of language itself. Its methods and findings inform research across multiple disciplines.

Language dating and prehistory

Linguistic evidence helps date events and migrations that left no written record. By reconstructing Proto-Indo-European vocabulary (which includes words for wheels, horses, and copper but not iron), researchers can estimate when and where the proto-language was spoken. Correlating linguistic evidence with archaeological findings strengthens theories about prehistoric population movements.

Linguistic archaeology

Reconstructed vocabulary serves as a window into prehistoric cultures. Proto-Indo-European contains words for specific animals, plants, technologies, and social roles, revealing details about the daily life of its speakers. Loanword analysis can trace ancient trade routes and cultural contacts; linguistic evidence of Silk Road interactions, for example, shows up in vocabulary shared across languages along the route.

Genetic vs. linguistic relationships

Comparing language family trees with genetic population data often reveals striking parallels, since languages and genes both spread through migration. But the two don't always match. Language shift can cause a population's genetic heritage and linguistic heritage to diverge: a community might adopt a new language while retaining its genetic profile. Studying these mismatches helps researchers build a more complete picture of human prehistory.

Challenges in historical linguistics

Every field has its limitations, and historical linguistics is no exception. Being aware of these challenges helps you evaluate claims about language history more critically.

Limitations of the comparative method

The comparative method requires enough data from related languages to identify patterns. For languages with few or no known relatives, it simply can't be applied. The method also works best for phonology and basic vocabulary; reconstructing syntax and semantics is much harder. There's also the challenge of distinguishing features that languages inherited from a common ancestor from features they developed independently or borrowed from each other.

Dealing with fragmentary evidence

Many ancient languages survive only in limited inscriptions or damaged manuscripts. Some writing systems, like consonantal alphabets that don't record vowels, leave significant gaps in our knowledge of pronunciation. Interpreting isolated words or short phrases without broader context is inherently uncertain.

Reconstructing proto-languages

The further back in time you go, the less certain reconstructions become. Features that weren't preserved in any surviving daughter language are essentially invisible to reconstruction. There's also a risk of unconsciously projecting modern linguistic assumptions onto ancient languages. Phonology and basic morphology can be reconstructed with reasonable confidence, but syntax and meaning are much harder to recover.