Idle thoughts upon the Ides of March: the feathered man

« previous post | next post »

It's a bad month in general:  dark, dreary, drizzly, dank, and damp.  Soon one's thoughts are flitting* about as though one had taken wings, like Eros or Cupid.

In Chinese mythology, there is a deity called Yǔrén 羽人 ("Feathered Man").  It has an ambiguous origin — first appears in Shānhǎi jīng 山海經 (Classic of the Mountains and Seas) and Chǔ cí 楚辭 (Songs of the South / Elegies of Chu), both circa mid to late 1st millennium BC.  Neither of these texts were in the Confucian mainstream, and in later times were relegated to an amorphous "Daoist" cultural current.

There are many early representations of Feathered Man".  If you want to get a good sense of what he looks like, here is a generous selection of images.

I note that "Eros" lacks a clear etymology.  Ditto for "Feathered Man".  I'm wondering if both of them could have emerged from that soup of Central Asian myth origins that Adrienne Mayor has previously often explored so fruitfully:  Amazons, fossils, poison weapons, tattoos, and so forth.

Chris Button notes:

Pulleyblank once speculated that 人 might originally have ended in -m. I'd have to think about the reconstruction a little more, but if we go with something like njǝːm, then we get ʁaːɣɁ    njǝːm, which isn't miles away from Greek éramai going back to PIE h1rem-. Just a thought.

I should note that Christopoulos Lucas is preparing a major paper on the Graeco-Sinitic connections between Eros and Feathered Man.


*From Chris Button's draft comparative etymological dictionary:

ʁaːʁɁ | cp. IE pləw- “flow” | see 永, 雨


羽 feather
羽 | •U (u); •ha, •hané

羽 | yǔ (yə ̯́ , wuəɁ̯ , ʁaːɣɁ) cp. plume (pləw-)


栩 waft, flit 栩 | KU (ku)

栩 | xǔ (xyə ̯́ , xuəɁ̯ , χaːɣɁ) cp. flit, flutter (pləw-)


Selected readings

Last moment addition from Adrienne Mayor, just as I was about to push the "Publish" button

Does the Feathered Man of the Classic of Mountains and Seas have a feathered body or is he simply winged? The only beings I know of in Greek mythology depicted with feathered bodies and wings are the female Harpies and Sirens. There are Greek artistic depictions of birds with women’s heads.

Eros—doesn’t his name derive from Greek for “sensual love/desire”? His parentage is, however, quite ambiguous: he is variously said to be the offspring of Ares god of war and Aphrodite goddess of love; or Ouranos and Gaia, or Ouranos and Aphrodites, or Zephyr (the West Wind) and Iris; Erebus (Darkness) and Nyx (Night), Aither (Brightness) and Nyx; etc. I think the earliest mention is Hesiod (ca 750 BC) who says Eros simply arose after Chaos. A Homeric Hymn refers to Eros as Pteros (Winged).

I believe his wings are generic signals for his semi-divine, supernatural nature as a primal force; he is often shown as a small winged figure hovering over lovers in early vase paintings (5th c BC). He’s never shown with a feathered body. So I’m not sure whether he can be related to the Central Asian Feathered Man imagery. I don’t know of any Scythian images of winged men.


Perhaps we should refer to Yǔrén 羽人 as "Winged Man" rather than as "Feathered Man".

Update from Adrienne Mayor: The feathered man's wings are sharp, reminiscent of some griffins from Minoan and Near Eastern art.


  1. martin schwartz said,

    March 11, 2023 @ 8:05 pm

    @Chris Button notes: ??? You are reconstructing the featherless Chinese word for 'man', yes? What does that have to do with the Greek verb éra-mai (-mai is 1st pers. middle ending) and its related noun érōs 'love'? Passing over the Sinito-Indo-European lubrications of lucubration, I note that the latter Greek forms cannot be from
    PIE *√h1rem 'to relax etc.'; for connection with the latter one would
    have to posit a more original (and really unsupported) arche-root
    ** √h2er and an extended form ***√h1erh2 alongside *√h1rem.
    Radically paraphrasing Emily Dickinson:
    Is love the man in feathers
    in dark and dreary weathers?
    I'f so, I'd keep the ****√s in tethers.

  2. Victor Mair said,

    March 11, 2023 @ 8:59 pm

    From Alan Kennedy:

    The history of the feathered cape is worth looking into.

    Female Immortal with Fruit and Feather Cape

    ca. 1500

    Davis Museum at Wellesley College

  3. Victor Mair said,

    March 11, 2023 @ 9:27 pm

    Selections from two magisterial Wikipedia articles on:

    Xian (Taoism)


    Victor H. Mair describes the xian archetype as:

    They are immune to heat and cold, untouched by the elements, and can fly, mounting upward with a fluttering motion. They dwell apart from the chaotic world of man, subsist on air and dew, are not anxious like ordinary people, and have the smooth skin and innocent faces of children. The transcendents live an effortless existence that is best described as spontaneous. They recall the ancient Indian ascetics and holy men known as ṛṣi who possessed similar traits.


    The etymology of xiān remains uncertain. The circa 200 CE Shiming, a Chinese dictionary that provided word-pun "etymologies", defines xiān (仙) as "to get old and not die," and explains it as someone who qiān (遷 "moves into") the mountains."

    Edward H. Schafer defined xian as "transcendent, sylph (a being who, through alchemical, gymnastic and other disciplines, has achieved a refined and perhaps immortal body, able to fly like a bird beyond the trammels of the base material world into the realms of aether, and nourish himself on air and dew.)" Schafer noted xian was cognate to xian 䙴 "soar up", qian 遷 "remove", and xianxian 僊僊 "a flapping dance movement"; and compared Chinese yuren 羽人 "feathered man; xian" with English peri "a fairy or supernatural being in Persian mythology" (Persian pari from par "feather; wing").

    Two linguistic hypotheses for the etymology of xian involve Arabic and Sino-Tibetan languages. Wu and Davis suggested the source was jinn, or jinni "genie" (from Arabic جني jinnī). "The marvelous powers of the Hsien are so like those of the jinni of the Arabian Nights that one wonders whether the Arabic word, jinn, may not be derived from the Chinese Hsien." Axel Schuessler's etymological dictionary suggests a Sino-Tibetan connection between xiān (Old Chinese *san or *sen) "'An immortal' … men and women who attain supernatural abilities; after death they become immortals and deities who can fly through the air" and Tibetan gšen < g-syen "shaman, one who has supernatural abilities, incl[uding] travel through the air". | | Peri

    n Persian mythology, peris (singular: peri; from Persian: پَری, romanized: parī, Persian pronunciation: [pæɾiː], plural پريان pariyān, Persian pronunciation: [pæɾijɒːn]; borrowed in European languages through Ottoman Turkish: پَری, romanized: peri) are exquisite, winged spirits renowned for their beauty. Peris were later adopted by other cultures. They are described in one reference work as mischievous beings that have been denied entry to paradise until they have completed penance for atonement. Under Islamic influence, Peris became benevolent spirits, in contrast to the mischievous jinn and evil divs (demons). Scholar Ulrich Marzolph [de] indicates an Indo-Iranian origin for peris, which were later integrated into the Arab houri-tale tradition.

    The Persian word پَری parī comes from Middle Persian parīg, itself from Old Persian *parikā-. In Persian language, the word 'Par', means Wing.

    The word has been borrowed in Azerbaijani as pəri, in Hindustani as parī (Urdu: پری / Hindi: परी) and in Turkish as peri.

  4. Chris Button said,

    March 11, 2023 @ 10:31 pm

    @ Martin Schwartz

    Yes it seems pretty unlikely, doesnt it? I actually wrote that "note" as a very hasty two-minute response to an email from Victor that was originally on a wholly different topic. I hadnt given it any real thought and didn't know it was going to appear here, but i don't have an issue with it either.

    As for Pulleyblank's tentative idea that 人 "person" perhaps had an -m coda, it's actually a very interesting discussion in itself. It totally flies in the face of accepted approaches (in true Pulleyblank style), but there is some tantalizing evidence. I'll try to post some here later if I have the time.

  5. martin schwartz said,

    March 12, 2023 @ 12:08 am

    There is no route whereby Arabic jinnī, pl jinn, woukd have come from Chinese vel sim. Such creatures can fly in fantasyover long distances; not so easily words. The point is that there are two
    etymological possibilities for the Arabic word: Syriac genyā,
    cf. Lat.genius? or the Semitic root g-n-n > Arabic j-n-n
    'to cover, enclose, benight'; hence majnūn 'insane'. It just occurred to me that that the latter has a parallelism of sorts to the etymology of Pers. parī, MPers. parīg < *parīkā. The latter has nought to do with Pers. par(r) 'wing, feather' < *parnah-, and the Middle Persian and Avestan evidence shows the critter to have been an unfairlylike
    demoness. In my study of the word in its older contexts, I
    derived OIr. parīkã < 'she who surrounds', as a succubus
    (female incubus). See my "Two Avestan demonesses…"
    in Bulletin of the Asia Institute 22 (2008).

  6. Chris Button said,

    March 12, 2023 @ 8:35 am

    On 人 "person" EMC (Early Middle Chinese) ɲin as possibly OC (Old Chinese) ᶮʝǝːm instead of OC nǝːɲ …

    人 as ᶮʝǝːm appears to be phonetic (based on oracle-bone forms) in 年 EMC nɛn, OC ᶮʝǝmː "year", which is clearly related to 稔 EMC ɲimɁ, OC ᶮʝǝːmɁ "(ripe) harvest", both related to Mon-Khmer cnam “year” (Pulleyblank notes the MK association, but I think Benedict might have been the first to link them in 1976).

    人 as ᶮʝǝːm also appears to be phonetic (based on oracle-bone forms) in 千 EMC tsʰɛn, OC cʰjǝmː ~ scjǝmː "thousand",† which Schuessler notes in his etymological dictionary to be a 'similar-looking word' to Mon lŋim "thousand", 'but the initial and final nasals do not agree with OC.' With the -m coda taken care of, my hunch here is that OC might have treated the l- prefix as something more like a lateral fricative ɬ- (ɬ ~ s alternations well-attested cross-linguistically, and OC sl- and ɬ were already emerging as allophones) to bring it more in line with the OC s- prefix (an l- prefix being impossible in OC). The onset ɬŋ- could then have ended up becoming associated with cʰ- ~ sc- as the coronal fricative combined with the velar nasal (compare how OC sŋ- ~ ʰŋ regularly gives EMC x- or when palatalized ɕ-)

    † On the broader phonology here, compare something like the relationship of 二 EMC ɲiʰ, OC ᶮʝǝːjs "two" with 次 EMC ʦʰiʰ, OC cʰǝːjs ~ scǝːjs "next, subsequent".

  7. Jerry Packard said,

    March 12, 2023 @ 11:45 am

    Literally none of the Chinese regionalects have -m as a coda for 人 ; it is either -n or -ng everywhere.

  8. Chris Button said,

    March 12, 2023 @ 12:11 pm

    The idea is that -jǝm (but crucially not -jǝmɁ) merged with -ǝɲ, which was originally -jǝŋ, very early on–hence there is no need to reconstruct a Shijing rhyme -jǝm unless evidence like the above is marshaled.

  9. Chris Button said,

    March 12, 2023 @ 12:12 pm

    But we do still need to reconstruct -jǝmɁ

  10. Feriaysh said,

    March 12, 2023 @ 12:25 pm

    Is the ancient Egyptian rishi (feathered) coffin image related to this discussion?

  11. Chris Button said,

    March 12, 2023 @ 12:27 pm

    Actually I should say “Old Chinese” rhyme -jǝm (I suppose that would be a Baxter 1992 -im) rather than “Shijing rhyme” because the Shijing rhyme is just “-ǝm” regardless. The “front vowel hypothesis” with Baxter’s -im as a separate rhyme is merely a surface phenomenon that does not affect underlying phonology.

  12. Chris Button said,

    March 12, 2023 @ 12:32 pm

    I should also add that -ǝɲ came from a merger of original -jǝŋ or -jǝn. The proposal is just to add -jǝm to that merger but only under certain conditions (i.e. no glottal catch)

  13. Victor Mair said,

    March 12, 2023 @ 12:33 pm

    From Lucas Christopoulos:

    What were the cultural environment, military power, and social system of some Saka Scythians, Sogdians, and other Bactrians from 250 BC to 100 BC in Central Asia, Eastern Central Asia, and then later during the Kushana Kingdom? What were the main mythological references and art innovations during that period in that area ?

    So far, Chinese scholars divide the representation of Yuren (Feathered / Winged Man) into: 1. That of the Shang dynasty and the early Warring States and 2. The one from the Qin-Han period. There are plenty of references and iconographical examples in my article to come : "Yuren, from Feather man to Eros," thus no need to elaborate here. What surprises me regarding the Shang dynasty Yuren, however, is the one depicted on the wine ewer in the photograph below. Does the face on that ewer appear to be Caucasian? What Indo-European tribe could have been represented there in 1200 BC?

    Ritual wine ewer (gong) with masks (taotie), dragons, and real animals – Smithsonian's National Museum of Asian Art
    Ewer with design that incorporates many stylized birds and animals. No inscription.

    VHM: If you go to the webpage of the Smithsonian Institution's National Museum of Asian Art for this piece (, you can view it from four different angles, close up in fine detail, and further away. The human figure is on the hind right leg, and a similar human figure is on the hind left leg. They seem to be wrapped in snakes and wear elaborate, ornamented helmets. The label says that the vessel was most likely made in southern China.

  14. Pamela said,

    March 12, 2023 @ 2:42 pm

    the Feathered Man figures look irresistibly like common depictions of Kokopelli.

  15. martin schwartz said,

    March 13, 2023 @ 1:09 am

    Human-haded winged lions are found throughout the ancient
    Near East. They figure on seals from the Holy Land, with headdresses pointing to Egyptain origin; these are the original
    cherubs Heb. k(e)rūß > Greek gryps (grúps) 'griffin'. There was
    an Irano-Caucasian winged griffin called paskuč (see Encyclopaedia
    Iranica online article "Sīmorg", which may be a Scythian name;
    appropriate critters are found in Scythian and Central Asian art.

  16. Philip Anderson said,

    March 13, 2023 @ 8:08 am

    Regarding ’winged’ v ‘feathered’, both ’pteros’ and ‘feather’ go back to the same PIE root (as do Latin ‘penna’ and Welsh ‘adar’ birds). Although we are accustomed to standard, classical representations of human/animal hybrids, earlier representations might mix the different components in a different way. I suspect that too much emphasis might be laid on the difference in English between winged and feathered.

  17. David Deden said,

    March 13, 2023 @ 11:34 pm

    man (n.)
    "a featherless plantigrade biped mammal of the genus Homo" [Century Dictionary], Old English man, mann "human being, person (male or female); brave man, hero…

  18. Chris Button said,

    March 14, 2023 @ 8:52 am

    Since 䙴 is in the bronze inscriptions and seems to predate 仙/僊 with a clear semantic association via 遷, etc. I wonder if an internal origin for 仙/僊 might be more appropriate? Any external links would then be coming from China rather than into China,

  19. Philip Anderson said,

    March 14, 2023 @ 8:53 am

    @David Deden
    Since there are no feathered mammals outside mythology, that definition seems overly detailed.

  20. Chris Button said,

    March 18, 2023 @ 6:06 am

    On the possibility of 人 "person" as ᶮʝǝːm, Pulleyblank tentatively suggests a possible connection with 男 nǝmː "man". The discrepancy between ᶮʝ- and n- supports a loanword origin (Schuessler's dictionary has some possible Monic comparanda with ɲ-)

RSS feed for comments on this post