Talk:Xiongnu language

Information from Xiongnu should be moved here

It seems to me that the rather extensive discussions at Xiongnu#Ethnolinguistic origins (which unfortunately conflates language and ethnicity) should be copied here and shortened at the other article. It should probably be subjected to a great deal of scrutiny and double-checking as well. I'm not sure all of it is sourced.--Ermenrich (talk) 14:39, 3 July 2024 (UTC)

Indo european

They were likely Indo-European. 2001:1C02:2C0D:100:18A2:AE66:D9B6:3C0 (talk) 13:23, 23 September 2025 (UTC)

Not according to the majority of scholars.—-Ermenrich (talk) 13:56, 23 September 2025 (UTC)

Xiongnu Language Classification: Reevaluating the Disregard for Turkic Evidence

I would like to open a discussion regarding the current presentation of the Xiongnu language classification. There is a significant discrepancy between the consensus—which often overlooks or marginalizes the Turkic hypothesis—and the body of historical and contemporary evidence linking the Xiongnu to the Turkic ethnogenesis. 1. Historical Continuity in Chinese Chronicles Chinese sources provide explicit connections between the Xiongnu and subsequent Turkic tribes:

The Weishu (103) explicitly states that the language of the Gaoche (Tiele) tribes was the same as that of the Xiongnu, with only minor differences, describing them as the "junior relatives" of the Xiongnu.
The Xin Tangshu (232) identifies the predecessors of the Huihe (Toquz Oghuz) as Xiongnu, noting their shared tradition of high-wheeled carts.
The Ashina clan, which founded the Göktürk Empire, is documented in the Book of Zhou and the History of the Northern Dynasties as a component of the Xiongnu confederation.
The Yueban (Örpün), described as remnants of the Northern Xiongnu, are identified by Chinese historians as an early Turkic tribe.

2. Onomastic Evidence The work of Otto Maenchen-Helfen, who compiled approximately 150 Hunnic/Xiongnu personal names, demonstrates a predominantly Turkic character. It is problematic that this extensive onomastic dataset is often sidelined in favor of studies that rely on an extremely limited selection of names (e.g., fewer than 10 entries) to propose alternative hypotheses like Yeniseian. 3. Genetic and Cultural Correlation Recent studies, such as the data presented by Akbari et al. (2026), show that medieval Anatolian Turks carry the Proto-Turkic/Xiongnu C-Y10420 paternal line. Furthermore, archaeological findings from the Ar Gunt cemetery link the elite Xiongnu genetic markers (N-M2019) to the European Huns. Conclusion Given this convergence of historical, linguistic, and genetic data, the current Wikipedia article seems to rely disproportionately on speculative and statistically weak models, while ignoring the explicit records of cultural and linguistic continuity found in primary Chinese chronicles and the foundational work of Maenchen-Helfen. Why does the article continue to present the Turkic hypothesis as merely "debated" when the primary historical records explicitly equate these groups with known Turkic predecessors? Squeaks 41 (talk) 17:29, 20 June 2026 (UTC)

Because the WP:RS/AC is not that the Xiongnu spoke Turkic. No one knows what language they spoke, and most scholars seem to lean against the Turkic hypothesis nowadays anyway.--Ermenrich (talk) 17:40, 20 June 2026 (UTC)

Your assertion that "no one knows" what language the Xiongnu spoke is a convenient dismissal that functions more as a tool for obscuring historical reality than an expression of academic neutrality. By framing the linguistic identity of the Xiongnu as a complete mystery, you effectively disregard the substantive, multi-disciplinary evidence that points toward a Turkic ethnogenesis, choosing instead to hide behind a bureaucratic definition of "academic consensus" that prioritizes trendy, speculative models over established primary data. It is a fundamental failure of the scientific method to elevate a 10-word analysis to the level of a primary hypothesis while simultaneously discarding the over 80 documented Hunnic and Xiongnu personal names compiled by scholars such as Maenchen-Helfen and Hyun Jin Kim, which exhibit clear and undeniable Turkic etymological correlates. This is not neutrality; it is selective censorship. The reliance on the most recent publication—such as the 2025 Fries and Bonmann study—as the definitive "consensus" ignores the fact that modern trends in academia often favor revisionist narratives that seek to "de-Turkify" Central Asian history, even when that history is explicitly documented in primary Chinese chronicles. When the *Weishu* and *Xin Tangshu* explicitly link the Gaoche, the Tiele, and the Ashina clan directly to the Xiongnu as their ancestral stock, you are not engaging in objective history, but rather participating in the systematic erasure of historical continuity. Furthermore, the dismissive attitude toward the genetic reality—such as the C-Y10420 and N-M2019 paternal lineages that trace a clear, empirical path from Xiongnu elites to later Turkic populations—reveals an underlying agenda that seeks to decouple culture and genetics to suit a preconceived linguistic theory. Wikipedia has become a repository for this kind of "trash" scholarship, where the platform's veneer of objectivity is used to shield ideologically driven editors from the obligation of engaging with the totality of the evidence. One must wonder why there is such a persistent, palpable animosity toward the Turkic hypothesis, even when it is supported by the convergence of historical records, onomastics, and genetic data. By insisting that your version of the "consensus" is the only one permitted on the page, you are not preserving the integrity of an encyclopedia; you are actively polluting it with modern, low-data speculation that ignores centuries of established historical record. The truth does not change simply because a new, flawed study is published, yet your editing practices ensure that Wikipedia remains a place where historical reality is discarded in favor of whatever currently suits the biases of a vocal academic minority.

It is remarkably telling that the Xiongnu language article omits these foundational Chinese historical records entirely, as if the primary witnesses to the Xiongnu's existence have nothing of value to say. Instead, you and your fellow editors have taken it upon yourselves to arbitrarily categorize the Xiongnu language as "Yeniseian," a leap of faith that defies both the historical texts you choose to ignore and the overwhelming weight of evidence. It is a laughable display of editorial arrogance to categorize a major nomadic language group based on the whims of a vocal minority while airbrushing away the primary sources that actually link these people to the Turkic world. Furthermore, the dismissive attitude toward the genetic reality—such as the C-Y10420 and N-M2019 paternal lineages that trace a clear, empirical path from Xiongnu elites to later Turkic populations—reveals an underlying agenda that seeks to decouple culture and genetics to suit a preconceived linguistic theory. Wikipedia has become a repository for this kind of "trash" scholarship, where the platform's veneer of objectivity is used to shield ideologically driven editors from the obligation of engaging with the totality of the evidence. One must wonder why there is such a persistent, palpable animosity toward the Turkic hypothesis, even when it is supported by the convergence of historical records, onomastics, and genetic data. By insisting that your version of the "consensus" is the only one permitted on the page, you are not preserving the integrity of an encyclopedia; you are actively polluting it with modern, low-data speculation that ignores centuries of established historical record. The truth does not change simply because a new, flawed study is published, yet your editing practices ensure that Wikipedia remains a place where historical reality is discarded in favor of whatever currently suits the biases of a vocal academic minority. Squeaks 41 (talk) 19:10, 20 June 2026 (UTC)

please see WP:original research. Beyond that, this is WP:NOTAFORUM, and a WP:WALLOFTEXT is not appreciated.—-Ermenrich (talk) 19:45, 20 June 2026 (UTC)

Please note that my previous contributions strictly adhere to WP:RS and WP:NPOV, providing primary historical documentation and peer-reviewed genetic data that you have yet to engage with. Dismissing substantive evidence as "original research" or "forum-like" when it contradicts your preferred narrative is a violation of WP:DISRUPT, not a defense of policy. You are here to build an encyclopedia based on the totality of evidence, not to curate a personal collection of low-data speculations while censoring foundational sources.

If your "academic consensus" cannot withstand the weight of primary historical records and modern data, perhaps it is time to reassess your role as an editor rather than an arbiter of history. Squeaks 41 (talk) 21:23, 20 June 2026 (UTC)

It’s not “my” academic consensus, it’s the consensus of actual scholars on the issue, I.e. not your personal opinion on the subject based on whatever research you have personally done.—-Ermenrich (talk) 22:15, 20 June 2026 (UTC)

Which scholars and academic consensus specifically? Squeaks 41 (talk) 05:49, 26 September 2026 (UTC)

Categorization of the Xiongnu language

I noticed that the Xiongnu language was categorized under Yeniseian languages ​​in the categories section, and I tried to have it removed, but I couldn't. Isn't this a complete disgrace in this consensus? We don't have definitive data on what language the Xiongnu spoke, and yet it's been placed in the Yeniseian category as if there were written evidence of the Xiongnu language (excluding words whose structure has been distorted by Chinese transcription). Squeaks 41 (talk) 14:03, 22 June 2026 (UTC)

Unclassified infobox, Yeniseian category: which is it?

The infobox on this article says the Xiongnu language is ‘’‘unclassified’’’. The same article is (or has been) placed in [[:Category:Yeniseian languages]] and/or presents a family tree with ‘’‘Xiongnu’’’ as a daughter of Southern Yeniseian / Pumpokolic. Those two things cannot both be true.

‘’‘1.’’’ Under [[WP:CAT]], membership is a classification claim, not a “maybe”. If the language is unclassified, on what policy basis is it in a family category at all? If the category stays, why does the infobox still say unclassified? Please pick one.

‘’‘2.’’’ [[WP:RS/AC]]: which ‘’‘secondary’’’ source says there is currently a scholarly consensus that Xiongnu was Yeniseian? Bonmann & Fries (2025) in ‘‘Transactions of the Philological Society’’ is a ‘’‘proposal’’’ (“suggests”, “almost certainly”, “tentatively… Old Arin”). A proposal is not AC. Why is a 2025 hypothesis doing work that the infobox itself refuses to do?

‘’‘3.’’’ The same page’s own “prevailing view” problem. Savelyev & Jeong (2020), [https://doi.org/10.1017/ehs.2019.15 ‘‘Evolutionary Human Sciences’’]: quote: the “traditional and ‘’‘prevailing’’’ view” is that the Xiongnu and/or the Huns were Turkic (Shiratori; Benzing; Tenišev; Schönig; Dybo 2007; Janhunen). They add that published Xiongnu genomes ‘’‘speak against’’’ the Yeniseian hypothesis if modern Kets are the proxy. Why does the article give Yeniseian a ‘’‘tree and a category’’’ and give the view these authors call prevailing a question mark in a list? That is the definition of [[WP:UNDUE]].

‘’‘4.’’’ Where is the Chinese testimony in the classification section? ‘’[[Book of Wei]]’’: Gaoche speech is essentially that of the Xiongnu, small differences. Later official histories: [[Tiele people|Tiele]] as Xiongnu remnants. Nobody treats Gaoche/Tiele as Yeniseian. Why is that primary-historical bundle worth less than three Hunnic names and one Jie couplet?

‘’‘5.’’’ ‘’[[Book of Han]]’’: ‘‘chengli gutu chanyu’’, glossed in-place: ‘‘chengli’’ = sky. Standard comparison is PT ‘’*teŋri’’ (Clauson; Dybo). One competing Yeniseian etymology for ‘’‘one word’’’ (Vovin) does not wipe a gloss + a title formula. Why is the sky-word treated as optional colour, while Arin hydronyms in Bonmann & Fries are treated as phylogenetic evidence?

‘’‘6.’’’ The [[Jie language|Jie couplet]] is ‘’‘one sentence’’’, from a ‘’‘subordinate group’’’, two centuries after the Xiongnu empire, with a published ‘’‘Turkic’’’ parse (Ramstedt; Bazin; von Gabain; Shimunek 2015) ‘’‘and’’’ a Yeniseian parse (Pulleyblank; Vovin). On what due-weight rule does that sentence classify ‘’‘Xiongnu’’’ as a whole as Yeniseian? If it does not, it cannot support [[:Category:Yeniseian languages]] on ‘’‘this’’’ article.

‘’‘7.’’’ Bonmann & Fries build the Hunnic half on ‘‘Attila’’, ‘‘Eskam’’, ‘‘Atakam’’. The bog-standard etymology of Attila is Gothic ‘‘atta’’ + ‘’-ila’’ (Maenchen-Helfen; same suffix as Wulfila, Totila). If three names + “almost certainly” can move a ‘’‘category’’’, why did Maenchen-Helfen’s much longer Turkic list (Dengizich < ‘’*deŋiz’’, Mundzucus < ‘’*munčuq’’, Oebarsios, ‘’-čur’’, ‘’-gur’’ tribal names) never justify [[:Category:Turkic languages]]? Asymmetric burden. Please justify the asymmetry in policy terms, not in “we must be careful with Turkic”.

‘’‘8.’’’ Double standard on certainty language. This talk culture objects to wording like “almost certainly” when it is used for Turkic. Bonmann & Fries use that wording for Arin and it is enough to shape categories. Either hedging applies to ‘’‘both’’’ hypotheses or it is not hedging, it is a veto on one side. Which?

‘’‘9.’’’ Genetics is not a language ID, but it is relevant to ‘’‘weight’’’ if 2025 linguistics is being used that way. Savelyev & Jeong: Xiongnu profile vs Ket proxy, against Yeniseian. Gnecchi-Ruscone et al. (2025): long IBD between high Xiongnu elite and some European Hun elites — i.e. the link Bonmann wants is genetic-historical, not Arin-specific. If 2025 papers can rewrite the tree, why do the aDNA papers that undercut a Ket-like core not even brake the category?

‘’‘10.’’’ Direct, falsifiable asks:

•  Remove [[:Category:Yeniseian languages]] from this article (and any cladogram that treats Xiongnu as a classified Yeniseian daughter) ‘’‘unless’’’ a secondary RS is produced that treats that classification as established.

•  Keep (Para-)Yeniseian, Turkic, Iranian, Mongolic in the infobox as competing hypotheses under ‘’‘unclassified’’’, which is what the lead already claims.

•  If editors believe the 2025 paper ‘’‘is’’’ enough for a family category, say so on-wiki as a [[WP:DUE]] claim so it can go to [[WP:CFD]] and [[WP:NPOVN]] instead of dying as a local deadlock.

This is not a request to make the infobox say Turkic. It is a request to stop giving one minority 2025 hypothesis a harder classification status than the article’s own “unclassified” label, and harder than the view Savelyev & Jeong call prevailing. If that is not [[WP:UNDUE]], what would be? Squeaks 41 (talk) 07:18, 25 September 2026 (UTC)

Is there a reason this WP:wall of text is all in bold?—-Ermenrich (talk) 15:11, 25 September 2026 (UTC)