Several dimensions, not one ladder
Traditional labels such as isolating, agglutinative and fusional can be useful first impressions, but WALS chapters 20 and 21 separate at least two questions that a single label conflates. Fusion concerns how closely a grammatical formative is phonologically joined to its host. Exponence concerns how many grammatical categories one formative encodes. A language can attach one marker per category, attach a marker carrying several categories, use a free word for one category, and change a stem internally for another. These are properties of constructions and markers, not stages of advancement.
An isolating marker is a phonological word on its own. A concatenative marker attaches to a host. A nonlinear marker may alter its host internally by vowel pattern or tone. WALS chapter 20 uses Fijian to show an independent past marker alongside a bound transitivity suffix in the same language. A single example thus defeats the idea that every marker in one language has the same fusion type.
One marker, one or several meanings
A monoexponential formative codes one category in the comparison at hand; a cumulative formative codes more than one. WALS chapter 21 cites Russian case-number endings as cumulative. This does not mean a learner can always split the ending into a “case part” and a “number part.” Nor does a separate suffix always carry exactly one function in every context. Syncretism can make one surface shape serve several paradigm cells, and an apparent segmentation may depend on historical rather than synchronic analysis.
WALS separately defines which case and tense-aspect-mood formatives were sampled. That design matters: different markers within a language can behave differently, and comparing the easiest marker in one language with the most complex in another would distort the result. A good teaching table records the whole attested form, proposed segmentation, category values, and the source for the analysis.
Ancient evidence
Do not combine a language’s inflectional forms from distant centuries into one timeless “type.” Written forms may obscure phonological words or tone, and editors may regularize spelling. A single ancient token cannot reveal whether a suffix is productive, optional, or restricted by genre. If the surviving material supports only case endings and no conversational syntax, describe the observed noun morphology and leave the rest open.
Exercise: select one marker from a published grammar. Test fusion by asking whether it forms an independent phonological word; test exponence by listing the categories it distinguishes; then test whether it varies by stem class. Write three separate answers rather than one “agglutinative/fusional” verdict.
Modern-language examples on this page clarify a linguistic pattern. They do not establish unattested sounds or forms for an ancient language. Follow the cited sources below and keep editorial readings separate from the surviving trace.