Locate the region before drawing a border
Semigallian communities occupied lands across parts of modern southern Latvia and northern Lithuania. The Lielupe and its tributaries are central to this geography. Guntis Zemītis’s National Encyclopedia entry describes a historical area whose extent changed; it should not be reduced to the boundaries of a present-day administrative region.[1]
Mägi’s 2013 review brings settlement and burial evidence together. It discusses cemeteries as clues to habitation, villages along the Lielupe river system, and prominent hillforts such as Mežotne, Tērvete and Dobele. It also notes how much the reconstruction of society depends on written sources and interpretation. A well-furnished grave or a large stronghold does not by itself supply a complete political constitution.[2]
A disputed district: Sidabrė
Ernestas Vasiliauskas’s 2005 chapter asks whether Sidabrė formed an independent Semigallian land or belonged to a larger neighbouring district. He compares written records, archaeological sites and later linguistic and geographical evidence. Sidabrė castle appears in records of 1289 and 1290, but the political territory associated with it has to be reconstructed.[3]
Vasiliauskas argues for a separate Sidabrė land. That conclusion is a historical hypothesis, not a surviving surveyed boundary. His use of several kinds of evidence makes the reasoning inspectable; it does not remove the need to test how far later dialects or later borders can illuminate the thirteenth century.[3]
War, migration and an unreliable number
Thirteenth-century warfare against the crusading orders is central to medieval accounts of Semigallia. Zemītis discusses resistance, the loss of strongholds and movement towards Lithuania. He also explicitly treats the chronicled figure of 100,000 migrants as exaggerated. It should not be reused as a measured population total, however often it appears in later retellings.[1]
When reading a campaign narrative, distinguish an account of a fortress being abandoned from an account of an entire region losing its population. Record the event described, who wrote about it and the geographical scale. Look for settlement evidence before extending the result to every village or later generation.
What happened after the campaigns?
Vasiliauskas’s 2009 study examines settlement from the fourteenth to the early sixteenth century. It addresses the long-running question of whether post-war Semigallia was an empty landscape. Combining archaeological and linguistic evidence, he argues that part of the population remained during the fourteenth century. He also identifies later settlement by people from other regions.[4]
The study distinguishes northern and southern patterns under different political and cultural conditions: the northern part under Livonian rule, the southern part within the Grand Duchy of Lithuania. This is a more differentiated account than either total continuity without change or complete depopulation followed by a new beginning. It does not tell us exactly how many households stayed in each locality.[4]
What survives of the language?
Semigallian is not documented by a surviving body of texts written in the language. Dini’s survey reconstructs features chiefly through names and possible traces in later Latvian and Lithuanian dialects. Such evidence can support linguistic comparisons, but it cannot supply a complete recorded conversation or a precise date when the last speaker died.[5]
A useful local study can begin with one stronghold and nearby cemeteries. Keep the medieval place-name, modern identification, excavation dates and competing interpretations together. Compare the history of the place with the changing use of the regional name, rather than assuming that one label always describes the same people, territory and institutions.
References
Sources accessed on 6 November 2026. The locators below identify the relevant passages.
Guntis Zemītis. zemgaļi. National Library of Latvia, National Encyclopedia. Edited institutional reference entry. Accessed 6 November 2026. Revised 2026-03-20.
Locator: Kopsavilkums and Apdzīvotības areāls, Lielupe basin and changing extent; updated 20 March 2026. Etnosa izzušana, thirteenth-century resistance, departures to Lithuania and explicit warning about the chronicle’s 100,000 figure. Source language: Latvian. English paraphrase.
Marika Mägi. 2013. Late Prehistoric Societies and Burials in the Eastern Baltic. Archaeologia Baltica, 19, pp. 177 to 194. DOI: 10.15181/ab.v19i0.281. Klaipėda University Press. Scholarly journal research article. Accessed 6 November 2026.
Locator: pp. 181 to 182, Semigallia, settlement pattern, cemeteries and hillforts; p. 177, methodological caution.
Ernestas Vasiliauskas. 2005. Ar buvo žiemgalių Sidabrės žemė? / Did Sidabrė land of the Semigallians exist? Praeities puslapiai: archeologija, kultūra, visuomenė, pp. 191 to 206. Klaipėda University Press. Scholarly archaeological book chapter. Accessed 6 November 2026.
Locator: pp. 191 to 193, historical problem and castle records; p. 206, English summary. pp. 204 to 206, concluding reconstruction and English summary. Source language: Lithuanian, with English summary. English paraphrase.
Ernestas Vasiliauskas. 2009. Žiemgalos apgyvendinimo raida XIV a.–XVI a. pradžioje / The Settlement of Semigallia in the 14th–the Early 16th Centuries. Istorija, 74, pp. 4 to 12. Vilnius Pedagogical University / Istorija; Vytautas Magnus University repository. Scholarly journal research article. Accessed 6 November 2026.
Locator: p. 4, question and evidence; pp. 10 to 12, conclusion and complete English summary. pp. 4 to 5 and 10 to 12, northern/southern models and settlement conclusions. Source language: Lithuanian, with English summary. English paraphrase.
Pietro U. Dini. 2014. Foundations of Baltic Languages. Eugrimas, Vilnius. Updated English edition, translated by Milda B. Richardson and Robert E. Richardson. ISBN 978-609-437-263-6. Scholarly monograph. Accessed 6 November 2026.
Locator: pp. 318 to 320, sections 5.6.4 and 5.6.5, sparse onomastic evidence, morphology, lexicon and proposed substratum.