The people beneath the hills, and the complicated history of a beautiful name
Etching and mezzotint from Liber Studiorum. Designed and etched by Turner; mezzotint by Thomas Goff Lupton.
The Metropolitan Museum of Art · Public domain.
There are people whose arrival you hear before you see them.
Footsteps in a passage. Gravel shifting under a boot. A gate opening somewhere behind you.
The people John Gregorson Campbell wrote about did not always arrive that way.
In the stories he gathered across the Scottish Highlands and Islands, a traveller might look up and discover that someone was already standing across a stream. A figure might enter without the ordinary warning of footsteps. Things could disappear before anyone realised another presence had been near.
Campbell had a striking English name for these beings.
The still folk.
It appears in his book, published posthumously in 1900, alongside two other descriptions: the “people of peace” and the “silently-moving” people. Campbell was a Gaelic-speaking minister who had spent three decades on the island of Tiree, and his book drew heavily on oral traditions gathered across the Highlands and Islands. 1
Cornell University Library / Internet Archive.
View original scan · Read transcription · Original scan
The phrase is immediately seductive.
It sounds gentle. Secretive. Almost pastoral.
It is none of those things quite so simply.
And its history begins with a problem: “the still folk” is not an uncomplicated literal translation of the Gaelic name.
A beautiful phrase, with a problem
The beings Campbell was describing were known, among other names, as the daoine sìth.
Daoine means people. The trouble begins with sìth.
In Scottish Gaelic, sìth certainly means peace. But the historical language around fairies, supernatural beings and the mounds associated with them is more complicated. Dictionaries preserve related but distinct senses around peace, fairy terminology and the hillocks or fairy mounds called sìthean. Alexander Macbain’s early twentieth-century etymological dictionary even places separate entries for sìth as peace and sìth in fairy terminology side by side. Historical Irish sources likewise distinguish peace from the supernatural mound, while also preserving wordplay between the two. 2
In other words, the language itself has been inviting interpretation for a very long time.
Campbell’s interpretation was one of them.
For him, the “stillness” of the fairy people was chiefly about how they moved.
Ordinary human movement leaves evidence. A floor creaks. A foot falls. Grass shifts. Someone approaching announces themselves through sound before they reach you.
Campbell’s fairies could evade that warning. They might cross a room or approach a traveller without the expected noise. Their movement could seem almost to glide. Stillness, in this sense, was not inactivity. It was motion without the ordinary proof of motion. 1
That is rather more unsettling than “people of peace”.
And even Campbell’s contemporaries were not entirely persuaded by the linguistic explanation. The connection between peace, fairies and mounds was disputed even around the time his book appeared. 3
The sources refuse to make the story as tidy as the English phrase does.
That may be one reason the phrase survives so well.
A world under the hill
Modern fairies have accumulated a great deal of baggage.
Tiny bodies. Wings. Sparkles. Benevolence. A general tendency to inhabit illustrated woodland.
Campbell’s people are much harder to standardise.
His own accounts disagree about something as elementary as their size. Some are extremely small. Others resemble human beings. Others again appear enormous. The contradiction bothered Campbell enough that he tried to reconcile it, but the disagreement in the stories is more interesting than the solution he imposed upon them. 1
They are also strikingly domestic.
They dress. Work. Borrow. Entertain. Keep livestock. Form households. Earlier Scottish accounts describe children, nurses, marriages and deaths. These were not empty hills inhabited by vaguely supernatural lights. They contained societies. 4
One story turns on a being who appears to be a child but, when alarmed that the fairy hill may be on fire, worries about the tools of his trade.
The detail is wonderfully mundane.
There is a smithy under the hill. 5
Elsewhere there are cattle, dairying, grain, loans and stolen produce. The supernatural world is not separated from ordinary Highland life by subject matter. It is built from many of the same concerns. 1
Animals need tending.
Food can run short.
Equipment matters.
Neighbours borrow things.
Neighbours also steal them.
This is perhaps what makes the old stories stranger than the decorative fairies that came later: the other world resembles this one closely enough to interfere with it.
The hills were not scenery
The people Campbell described did not live in a generic enchanted landscape.
They were associated with particular ground.
The Gaelic sìthean can refer to a small hill or knoll and to a fairy hill. Campbell himself describes green, rounded elevations associated with fairy habitation. But the language needs care here too: a fairy mound should not automatically become an ancient burial mound every time the story is retold. A hill can be natural, archaeological, named, storied—or some combination of these—and a dictionary cannot tell us what lies beneath the soil. 6 7
What matters is that the stories existed within an inhabited landscape.
A person walked past a particular knoll.
Cattle grazed near it.
A mill stood somewhere nearby.
Music was heard from one hill and not another.
This changes the character of the folklore.
The fairy world was not necessarily far away. It could occupy a feature visible from home.
A green rise in the ground could remain entirely ordinary in daylight and still carry a story about what someone once heard there after dark.
The boundary between worlds did not always require a journey.
Sometimes it required only a different explanation for the same place.
A noisy people called still
There is another complication.
The still folk were not particularly silent.
Campbell records music. Singing. Speech. Shrieks. Rustling. Sounds resembling livestock. 1
Stories from elsewhere describe tunes heard coming from fairy dwellings, music passing into the human world even when the musicians remained unseen. 8
So still cannot sensibly mean mute.
Campbell’s idea was narrower: their approach could escape notice.
A busy household could exist under the hill while an individual from that household might arrive beside you without the warning you expected.
The distinction matters.
Stillness here is not serenity.
It is not quiet living.
It is the unsettling possibility that something has entered the field of attention before you have registered its arrival.
The phrase becomes better, not worse, once that distinction is restored.
Good neighbours
People have often given dangerous or unpredictable beings flattering names.
Campbell records another expression, daoine còire, which he renders as “honest folk”. He explains the courtesy practically: speaking badly of supernatural neighbours who might be listening was unwise. He compares the idea with the Irish “good people” and the Lowland Scottish “good neighbours”. 9
The compliment therefore tells us surprisingly little about character.
Calling someone good may be an assessment.
It may also be diplomacy.
And diplomacy was useful because the stories are morally unstable.
The fairy people may lend something, repay something or confer a benefit.
They may also steal livestock, take produce, abduct a person or substitute a changeling.
Campbell’s own material defeats any attempt to classify them neatly as benevolent or malevolent. One account may involve assistance while another involves profound danger. Even his contemporary reviewer noticed that Campbell sometimes tried to formulate rules that his own collection contradicted. 10
This is another reason “people of peace” should not be read as a character reference.
A courteous name is not the same thing as a safe neighbour.
The disturbing familiarity of the other world
Some of the darkest stories concern childbirth and children.
Campbell records fears surrounding women after birth, infants taken away, and changelings supposedly left behind. Other collections contain similarly disturbing material.
These stories are easy either to sanitise or sensationalise.
Neither is useful.
They belong to a world in which illness, bodily difference, difficult infancy, loss and unexplained behaviour could sometimes be interpreted through supernatural interference.
The stories could themselves prescribe violence against supposed changelings. 11
The point is not that everyone in Gaelic-speaking Scotland shared one fixed creed. Campbell explicitly warned that the traditions he assembled were not known uniformly in every locality. 1
Nor should folklore be reverse-diagnosed from a distance.
The value lies in seeing the explanatory world the stories made available.
A child who seemed profoundly changed might not simply be a changed child.
Someone else might be there instead.
The other world could enter the household by resembling its members closely enough to take their place.
Again, resemblance is more unsettling than difference.
Scotland, Ireland and the temptation of “Celtic fairies”
Scottish Gaelic and Irish traditions genuinely are related.
Their languages preserve closely connected vocabulary around fairy mounds and supernatural inhabitants. Irish aos sí, for example, belongs to the same broad linguistic world. 12
But similarity invites another kind of simplification: everything becomes “Celtic folklore”.
That label is often too large to be useful.
An Irish dictionary can illuminate the history of a related word. It cannot tell us what a nineteenth-century storyteller on Tiree meant in a particular story. Shared vocabulary does not make every being, custom or narrative role identical across centuries and regions. 12
There are differences within Scotland too.
Highland and Lowland traditions cannot simply be poured into one national container. Language matters. Place matters. Period matters. Who recorded the story matters. 13
The more closely we look, the less helpful the single generic fairy becomes.
That is not a disappointment.
It is where the subject begins to become interesting.
What survives
Folklore archives offer a peculiar kind of afterlife.
They do not tell us that an old belief survives unchanged.
They preserve occasions on which somebody said something.
A story told to a collector.
A tune recorded from a singer.
A place name written down.
A memory repeated years later, perhaps differently from the first telling.
One of the strengths of the surviving Gaelic material is that, sometimes, we can move back toward those individual occasions: a named narrator, a date, a locality, the Gaelic words themselves. 8
That is a better route into cultural memory than pretending there was once a single, coherent “fairy faith” shared everywhere and then simply lost.
The archive is messier.
So was the tradition.
And the phrase “the still folk” survives best when we allow it to remain slightly messy too.
Campbell used it.
He explained it through extraordinarily quiet movement.
His explanation was contestable even in his own period.
Behind it lies a cluster of Gaelic language around peace, supernatural beings and the mounds associated with them.
We do not need to resolve all of that into one neat translation.
In fact, doing so would remove exactly what makes the phrase worth investigating.
From the Still Folk to Stillfolk
Stillfolk is a modern name.
It does not claim descent from the daoine sìth, ownership of Gaelic tradition, or authority to speak for Gaelic communities. The historical phrase was encountered as an old resonance and admired for its unusual combination of quietness and ambiguity.
That resemblance belongs to the present.
It would be a mistake to turn Campbell’s interpretation into an ancient philosophy of childhood, or to treat Gaelic fairy belief as though it had somehow been waiting to explain a contemporary children’s brand.
The old stories deserve better than that.
They belong to particular languages, landscapes, tellers and records.
Our Stillfolk belongs to another story.
But there is something in the old English phrase that remains difficult to forget.
A presence that does not necessarily announce itself.
Something close enough to alter the ordinary world before anyone has quite noticed it arrive.
The still folk.
Sources and notes
Campbell, John Gregorson. 1900. Superstitions of the Highlands and Islands of Scotland: Collected Entirely from Oral Sources. Glasgow: James MacLehose and Sons. Editorial note and preface; chapter I, especially pp. 1, 3–4, 7–9, 11–14 and 15–35. Searchable text. The editorial note dates his Tiree ministry to 1861–1891; the book appeared posthumously. Page 1 supplies the three English descriptions, pp. 3–4 the account of movement and sound. This establishes a documented use, not that Campbell coined the phrase.
Macbain, Alexander. 1911. An Etymological Dictionary of the Gaelic Language. Stirling: Eneas Mackay, p. 325, separate entries for sìth as peace and in fairy terminology. Page transcription and scan. Also eDIL: An Electronic Dictionary of the Irish Language (2019), based on the Royal Irish Academy’s dictionary (1913–1976), s.v. ‘1 síd, síth’ (fairy mound) and ‘2 síd, síth’ (peace), including the wordplay cited as LU 9998–9. Dictionary. Scottish and medieval Irish evidence are distinguished; dictionary-mediated wordplay is not a new translation of the manuscript.
Nutt, Alfred. 1900. Review of Campbell’s Superstitions of the Highlands and Islands of Scotland. Folklore 11(4): 422–423, especially p. 423 on the peace interpretation. Page 422 · Page 423. Nutt disputes the etymology while acknowledging its apparent antiquity in Gaelic usage. The linked transcriptions are associated with scans and labelled as needing proofreading.
Kirk, Robert. 1893 [account composed 1691; first printed 1815]. The Secret Commonwealth of Elves, Fauns, & Fairies, introduction and notes by Andrew Lang. London: David Nutt, numbered sections 1–4. 1893 text. Read alongside Campbell, chapter I. Kirk’s account is distinguished from Lang’s later introduction and commentary.
MacDougall, James. 1910. Folk Tales and Fairy Lore in Gaelic and English, Collected from Oral Tradition. Edited by George Calder. Edinburgh: John Grant, pp. 100–103, the Torr-a-Bhuilg tale. University of Toronto digitised volume. This is the source of the apparent child concerned about his smith’s equipment, rather than an anecdote attributed indiscriminately to Campbell. Calder’s editorial and translation interventions should be borne in mind.
Dwelly, Edward. 1901–1911. The Illustrated Gaelic–English Dictionary, s.v. sìthean. Also Am Faclair Beag, s.v. sìthean. Search both dictionaries. The historical Dwelly entries and modern Am Faclair Beag results are distinct resources on the same website. See also Campbell, pp. 11–14.
Evemalm-Graham, Sofia. 2023. ‘The Fairy Dwellings of Iona: What Do the Place-names Tell Us?’ Iona’s Namescape, University of Glasgow, 18 May. Research article. Supports care over the relationship between names, tales, particular sites and archaeology; a fairy place-name alone does not establish a burial mound.
Wiseman, Andrew. 2013. ‘General MacDonald: A Fairy Tune.’ The Calum Maclean Project, 6 August. Transcript and commentary. Primary material: Peter MacCormick of Hacklett, Benbecula, recorded by Calum Maclean, 4 April 1950; National Folklore Collection, University College Dublin, NFC 1181, pp. 114–115. The project also discusses a 1953 telling. The original research consulted the transcripts, not the associated audio recordings. This is an example of music, named narrators and variation, not evidence for a single unchanged belief system.
Campbell, Superstitions (1900), p. 9, daoine còire and courteous names. Full text. The explanation of appeasing names is clearest here; it should not replace Campbell’s separate acoustic explanation of ‘still folk’.
Campbell, Superstitions (1900), pp. 22–24, on fairy gifts and a returned loan; Nutt, review (1900), p. 422. Nutt’s discussion. Nutt expressly notices the inconsistency between a general warning and the beneficial meal supply.
Campbell, Superstitions (1900), pp. 36–39, on childbirth and changelings, and preface on uneven local knowledge; MacDougall, Folk Tales and Fairy Lore (1910), pp. 142–147, the Rannoch tailor narrative. Campbell text · MacDougall volume. These are records of tales, reported beliefs and treatments. They do not independently establish that a particular historical child was harmed or license retrospective medical diagnosis.
Ó Dónaill, Niall. 1977. Foclóir Gaeilge–Béarla. Dublin: An Gúm, s.v. sí, including aos sí. Entry, Teanglann / Foras na Gaeilge. This is Irish lexicography, used for comparison rather than as authority for Scottish Gaelic usage. See also the separate historical Irish entries in note 2.
MacInnes, John. 2005. ‘Fairies and Folk.’ Textualities. Critical essay. Questions treating Highland Gaelic and Lowland traditions as culturally homogeneous while acknowledging shared material. Its other speculative arguments are not adopted here.