Two Ways to Classify a Language
When linguists classify languages, they use two very different approaches: genealogical and typological.
Genealogical classification asks where did this language come from? — grouping languages by shared ancestry into families and branches.
Typological classification asks what does this language look like? — grouping languages by shared structural features, regardless of origin. The results can be surprising: unrelated languages often converge on the same solutions, while close relatives can end up structurally very different.
Typological Categories
Word Order
One of the most studied typological features is the basic order of Subject, Verb, and Object in a sentence. English is SVO: The cat (S) ate (V) the fish (O). But languages vary:
- SOV — the most common type globally: Japanese, Turkish, Hindi (The cat the fish ate)
- SVO — English, Mandarin, Swahili (The cat ate the fish)
- VSO — Classical Arabic, Welsh, Tagalog (Ate the cat the fish)
- VOS, OVS, OSV — rare, but they exist
Morphological Type
Languages also differ in how they build words:
Analytic (isolating) languages use separate words for each meaning unit. Mandarin is the classic example — grammatical relationships are shown through word order and particles, not word endings.
Synthetic languages pack multiple meanings into a single word through affixes. English is mildly synthetic: walks bundles verb + third person + singular + present.
Agglutinative languages take this further with long chains of distinct, separable suffixes. Turkish evlerimden means "from my houses": ev (house) + ler (plural) + im (my) + den (from). Each piece does exactly one job.
Fusional (inflectional) languages merge multiple meanings into a single ending. Latin amō ("I love") fuses person, number, tense, and voice into one suffix that can't be broken apart.
Tone
About half the world's languages are tonal — pitch carries meaning at the word level. In Mandarin, mā (flat tone) means "mother" while mǎ (falling-rising tone) means "horse." These are entirely different words that happen to share the same consonants and vowel.
Tonal languages are especially common in:
- East and Southeast Asia (Mandarin, Cantonese, Vietnamese, Thai)
- Sub-Saharan Africa (Yoruba, Zulu, many Bantu languages)
- Parts of the Americas (many indigenous languages)
Why Typology Cuts Across Families
Genealogy and typology often diverge in interesting ways.
English and Mandarin are in completely different families — Germanic versus Sino-Tibetan — but both are analytic SVO languages. Meanwhile, English and German are close cousins (both Germanic) but German has held onto far more inflectional morphology than English has.
This happens because typological features spread through contact. Languages spoken in the same region for centuries tend to converge on structural features even without shared ancestry — a phenomenon linguists call a Sprachbund (language league). The Balkans are a famous example: Greek, Romanian, Bulgarian, Albanian, and Macedonian come from different branches but share a striking set of features found nowhere else in Indo-European.
Reading the Map
The language map on this site is primarily genealogical — coloured by family and branch. But typological patterns have their own geography. Tonal languages cluster in particular regions. SOV languages dominate a great arc from Turkey through India to Japan. These patterns are the fingerprints of ancient contact, migration, and convergence — a layer of history that genealogy alone can't show.