No other piece of iron turns up in Germanic graves as often as the seax, and none is as thoroughly misunderstood. It is knife and weapon at once, grave good and tool, status symbol and meat cleaver. It dates cemeteries to within a quarter of a century, it carries the only complete runic alphabet from England, and in the scholarly literature it is almost never called what half the reenactment scene calls it. A deep dive into a blade far older than the Vikings — and far more interesting.
A seax is, put briefly, a single-edged blade with a straight or angled back, forged in one piece, with a tang instead of a sword hilt. So far, so simple. The trouble starts immediately afterwards: where does the knife end and the weapon begin? Archaeology's answer is bluntly pragmatic — it draws the line in centimetres. Anything under roughly 20 centimetres of blade counts in most studies as a working knife; anything above gets counted, measured and sorted into types.
That arbitrariness is not sloppiness but self-defence. The transitions are fluid, and the category runs from kitchen blade to half a sword. Anyone talking about the seax is therefore never talking about one object but about a family of objects that changed steadily over some four hundred years. And which, precisely for that reason, can be dated beautifully.
Behind every form stands a Proto-Germanic word, reconstructed as *sahsa-, with the plain meaning “knife, short sword”. It survives in Old High German sahs, Old English seax, Old Norse sax, Old Saxon and Old Frisian sax — Saterland Frisian Saaks for “knife” is still in use today.
The word has kept its best foothold in modern Scandinavia, though with a shift in meaning that every reenactor eventually trips over: Swedish sax, Danish and Norwegian saks, Finnish (borrowed) sakset today simply mean “scissors”. The warrior's blade has become a household item — the double edge, in the figurative sense, remains. German long hesitated over the word's gender: Adelung still lists it as neuter in 1811, while modern dictionaries make it masculine.
The obvious idea is also the oldest: the Saxons are named after their weapon, they are the “knife people”. Widukind of Corvey told that story in the tenth century in his Deeds of the Saxons (Res gestae Saxonicae I,6–7): at a peace negotiation with the Thuringians the Saxons supposedly hid knives under their cloaks, cut down their unarmed counterparts and took their tribal name from the deed.
That is a good story. It is not an etymology. Widukind writes roughly five hundred years after the events he describes, and he writes as court historian to a Saxon dynasty that needed a glorious prehistory. Above all, though, the tribal name is markedly older than the weapon it is hung on. Saxones already appears in the Geography of Claudius Ptolemy in the second century AD — at a time when the Merovingian narrow seax did not yet exist. To name the Saxons after the seax is to name a people after a weapon that only became fashionable three centuries later.
This does not settle the matter entirely. Single-edged knives called sahs existed long before the Merovingian seax, and researchers such as Albert Genrich and Manfred Rech consider it possible that the name goes back to warrior groups in the Chaucian orbit who were labelled after their principal weapon. Possible — but unproven. Stated cleanly, the position is this: both words exist, a connection is conceivable, the direction of derivation is open, and Widukind's tale documents only what people in the tenth century would have liked to believe. More on that people in our overview of the Saxons.
In museums, catalogues and online shops the long seax is happily called a “scramasax”. It sounds like technical terminology and is in fact a single attestation. Gregory of Tours, in his History of the Franks (Historia Francorum IV,51), describes the murder of King Sigibert I at Vitry in 575 — and in doing so supplies the only ancient occurrence of the word:
Tunc duo pueri cum cultris validis, quos vulgo scramasaxos vocant, infectis veneno, maleficati a Fredegunde regina, cum aliam causam suggerire simularent, utraque ei latera feriunt.
“Then two young men struck him in both flanks with strong knives, the kind commonly called scramasaxes, smeared with poison — suborned by Queen Fredegund and pretending they had come to raise some other matter.”Gregory of Tours, Historia Francorum IV,51 (MGH SS rer. Merov. 1,1); translation: Glanz & Gravur
A second, later reference stands in the Liber Historiae Francorum (c. 727, ch. 35), where the same killing is described with duobus scramsaxiis — a variant of the same word, from the same narrative tradition. That is the entire source base. What scrama- actually means remains unresolved; readings from “wound” to “sharp” compete and none convinces. Recent scholarship therefore largely avoids the term and speaks of the long seax, or simply the seax. And beware the chronology: Gregory's murder happens in 575, decades before the age of the true long seaxes. He almost certainly means substantial fighting knives, not the late giants sold under the name today.
Before the types, a word on construction, because everything else hangs on it. Scholarship separates two groups. The older, pre-Merovingian seaxes have a back-set tang: the blade's back runs into the tang in one unbroken line, with grip plates riveted on at the sides — essentially a large knife.
Merovingian seaxes, by contrast, have a centrally set tang carrying a slid-on wooden grip. The tang is narrower than the blade and seats against shoulders on both back and edge. That sounds like a detail; it is the real break. An enlarged knife becomes a weapon grip that survives hard blows because the blade shoulders absorb the force. From here on the seax is not a big knife but a category of its own.
Oddly, the Merovingian sequence opens with its longest form. The “long narrow seax” of the fifth to early sixth century has blades over 30 centimetres with a width well under ten per cent of the length — an extremely slender object. It is found mainly in the middle Danube region and in what is now south-western Germany, occasionally among the Franks and in Gaul.
Dieter Quast has proposed tracing the form back to steppe-nomadic models transmitted via Byzantium to the Alemanni. Among the most prominent pieces are a scabbard mount from the grave of Childeric I and a gold grip ferrule from the princely burial at Ailenberg. Important for understanding what follows: this early narrow long seax is not regarded as the ancestor of the later series. It is a dead end with splendid representatives — the actual line begins smaller.
The short seax, at 28 to 30 millimetres wide, has an average blade length of 200 to 250 millimetres. The blade is plain and knife-like, the point sits between the centre and back lines, the short tang is centrally set, metal fittings on the wooden grip are rare. These blades develop out of late antique knife forms, and around the middle of the fifth century they reach their widest distribution in the Frankish area — they are still absent in northern Germany at this stage.
The short seax was, on current reading, handled like a dagger: thrust rather than cut. That fits Gregory's murder scene as well as the observation that short seaxes usually lie in graves as supplementary armament beside a sword, not as a main weapon.
From the middle of the sixth century the blades grow to an average of 300 millimetres in length and around 30 millimetres in width. The tang becomes long, seats on blade shoulders, and the grip now often carries a grip plate, a flush guard and a pommel. Overall length reaches up to 400 millimetres. Narrow seaxes now appear in northern Germany too.
From the last quarter of the sixth century comes Schmalsax II — and it is, against expectation, more plainly made than its predecessor. An average blade length of 290 millimetres, around 33 millimetres wide, point on the centre line, shorter tang stumps, no decoration. Such retreats in effort are typical of the category: the seax becomes more common, more widely distributed and therefore more everyday. What everyone owns need no longer shine.
At the end of the sixth century the proportions tip. The light broad seax averages 305 millimetres in length at 43 millimetres in width, the heavy broad seax 355 millimetres at around 49. The centrally set tangs frequently match the length of the blade. Blades carry decorative grooves and rectangular fullers following the line of the back; distinctive forging techniques are largely absent, but some broad seaxes carry inscriptions in runes or Latin letters.
Above all, though, the scabbard now becomes the showpiece — more on that shortly. The pommels are striking: from this phase they often bear bold three-dimensional decoration — animal forms, bearded faces, occasionally phallic motifs that appear to go back to late Roman models current in Gaul. Broad seaxes were frequently worn as a secondary weapon alongside the spatha and were probably handled like a machete: cutting, not thrusting.
From the last quarter of the seventh century the blades grow slimmer again and markedly longer — an average of 490 millimetres at around 42 millimetres wide, and in the final phase 520 millimetres on average. Refined forging and pattern welding keep that length light enough; the balance point moves close to the blade shoulder and the weapon handles like a sword. In the second half of the eighth century the point migrates towards the edge line.
The regional split is notable. In the Saxon-influenced landscapes — Westphalia, Lower Saxony — the long seax appears as the sole main weapon, without a spatha. In Frisian and Frankish-influenced regions it is more simply made and accompanies the double-edged sword as a side arm. Two societies, two philosophies of armament, the same blade. How it fits into the rest of the kit is set out in our weapons guide.
The sequence Kurzsax – Schmalsax – Breitsax – Langsax did not fall from the sky. Its foundation was laid by Kurt Böhner in 1958 with his study of the Frankish antiquities of the Trier region, the first large chronological analysis of a Frankish find area. Böhner's scheme has been corrected many times over — he is today, for instance, the only major researcher who still has the long seax in use into the ninth century, while the majority stop at the eighth at the latest.
The system was refined by Wolfgang Hübener in 1988, by Jo Wernard's study of the single-edged sword of the Merovingian period in southern Germany (Germania 76, 1998), and by Frank Siegmund's “Merowingerzeit am Niederrhein” (1998). The most frequently cited synthesis, with the measurements given above, is by Herbert Westphal in the Reallexikon der Germanischen Altertumskunde (vol. 26, 2004). So when someone says “after Böhner”, they almost always mean a system that has been worked on for four decades since.
Why all this fuss over millimetres? Because the seax is one of the best dating instruments of the Merovingian period. It changes fast, it lies in graves frequently, and it can be measured. Combined by seriation and correspondence analysis with belt sets, brooches and beads, it produces a fine chronology that pins individual burials to spans of sometimes only fifteen to forty-five years.
Siegmund did exactly that for the Lower Rhine: eleven Lower Rhine phases between the middle of the fifth and the middle of the eighth century, based on 316 male and 215 female graves with 158 and 124 types respectively. Together with the south-west German fashion phases, this became the basis of today's south German chronology. A seax is therefore less a weapon than a timestamp — just one you can pick up.
For a workshop near Aachen this is not distant material. The Lower Rhine is among the most densely researched seax landscapes in Europe, and the cemetery of Krefeld-Gellep, Roman Gelduba, is its centre: over 6,000 excavated graves, a sequence running from the first to the eighth century and thus one of the rare continuities from Roman into Merovingian times.
Excavation began in the 1930s under Albert Steeger, a geologist and headmaster who gave up teaching in 1936 to devote himself entirely to the dig; after his death in 1959 Renate Pirling took over and in 1962 uncovered the richly furnished grave 1782. Add the neighbouring territories — the lands of the Franks between Meuse and Rhine — and you have a body of material in which the whole development of the seax can be read, barely a hundred kilometres from our bench.
Look only at the blade and you miss the artistry. On broad seaxes it is the scabbards that carry the effort: leather over wood, bound with mouth plates and chapes, studded with rows of bronze rivets and bosses, plus punched or cut leather decoration. Original mouth plates are bronze and show motifs such as triskelia or animal whorls ending in bird heads.
The seax was not slung from a baldric but worn roughly horizontally on the belt, across the belly or hip, secured by several suspension loops and leather straps. That makes sense: a weapon you can reach in a crush and while seated, without half a metre of blade swinging near the ground. With the later long seaxes, incidentally, the effort swings back — their scabbards are conspicuously plain compared with those of the broad seaxes, often reduced on the reverse to small iron suspension loops. The display migrates from the scabbard into the blade.
The seax was long dismissed as a crude chopping knife of mediocre iron — the plain cousin of the noble sword. That verdict has been overturned. In a project run by Stefan Mäder in 1999, seax blades were polished by the Japanese specialist Sasaki Takushi and assessed by the swordsmith Amada Akitsugu, with startling results.
One examined blade from Bad Krozingen, dated between 550 and 650, proved to be a composite of core steel and jacket steel with differing carbon contents. The core steel had been folded roughly ten times, the jacket steel was homogeneous at around 0.5 per cent carbon. The blade had been selectively hardened — martensite crystals were detectable in the edge zone, probably produced with an insulating clay layer — and then tempered. This is not rough smithing; it is controlled heat treatment. How such work is done in practice is described in our article on smithing.
The question is usually posed as an either/or, and it cannot be answered that way. A short seax is a secondary weapon and an everyday implement at once; a heavy broad seax is beyond doubt a weapon; a narrow long knife of the Scandinavian kind was, judging by frequency and shape, more a slaughtering and butchering knife than a fighting blade. Rosalind Gale said precisely that about the Anglo-Saxon material back in 1989.
One find makes the case rather neatly: the long knife from the Oseberg ship burial lay together with kitchen equipment. Conversely, the ordinary working knife, Old Norse knifr, is the single most common metal find of the Viking Age and was worn on the belt by everyone, women and men alike. Between that everyday knife and the double-edged sword lies the whole spectrum, and the seax fills the middle. The boundary was blurred at the time, and the split into “weapon” and “tool” is our category, not theirs.
The most famous single piece of the whole category was found in 1857 by the labourer Henry J. Briggs in the Thames at Battersea: the Seax of Beagnoth, now in the British Museum (inv. 1857,0623.1). Total length 72.1 centimetres, of which 55.1 centimetres is blade and 17 centimetres tang, weighing 985 grams, dated to the tenth century. Twisted wires of copper, brass and silver are hammered into cut grooves, along with inlaid triangles and lozenges — on one side a lozenge pattern that may be meant to imitate pattern welding.
On the other side stands what makes the piece unique: the complete Anglo-Saxon rune row with all twenty-eight characters — the only known epigraphic attestation of the full futhorc — and, separated by a herringbone panel, the man's name Beagnoþ. The row is, however, faulty: runes 20 to 23 appear in an order attested nowhere else, the small s looks squeezed in as an afterthought, and several letterforms are unusual. The craftsman was evidently no practised rune-carver and probably copied from a manuscript.
Who Beagnoth was remains open. Raymond Page lists four possibilities — smith, rune-master, owner or giver — and closes with the most honest sentence in the literature: we cannot decide. Clearer is the seax from Sittingbourne in Kent, a ninth-century broken-back piece that names both roles in insular majuscules: “Biorhtelm made me” and “Sigebereht owns me”. Two names, two roles, one blade. Anyone thinking about engraving today will find the oldest version of the subject here — more in our overview of runes.
Now the awkward part. In popular culture the seax counts as a Viking weapon — and that is exactly where the evidence is thinnest. Tomáš Vlasatý has reviewed the Scandinavian material: in Denmark a grand total of four long knives of the relevant type are known, three from Hedeby and one from Skåne, against well over a hundred sword finds. In Norway about twenty pieces can be placed securely after 800, mostly 20 to 35 centimetres long and 2 to 3 centimetres wide. On Gotland there are 18 long knives — out of some 800 knives in total.
The only real flourishing is a tenth-century fashion in central Sweden, where at least 54 long knives are known, many with elaborately decorated scabbards enclosing blade and grip; to draw them at all, a small ring with a leather strap sat at the end of the handle. The design language of those scabbards is Baltic-influenced. What actually succeeds the seax in Scandinavia is not more seaxes but single-edged swords: Jan Petersen counted 370 of them in 1919 alone, roughly 23 per cent of all Norwegian swords he recorded. By the middle of the tenth century these too give way to the double-edged blade.
On the Continent the story ends in the ninth century — but be careful with that conclusion. What ends at that point is first of all the custom of furnished burial: the weapon graves stop, not necessarily the weapons. That the long seax remained in use for some time longer is suggested, for instance, by an image in the Stuttgart Psalter (WLB Cod.bibl.fol.23, fol. 5v, c. 820/830).
The real reason probably lies elsewhere. The Carolingian reform period saw central workshops producing double-edged swords in previously unknown numbers and quality, with better blades and more advanced hilts. Against that revolution in armament the seax looked archaic. On the British Isles and in Scandinavia it held on for a while — as a tool, as a status object, as regional fashion. But the main age of the seax lies in the sixth to eighth centuries, and by then it was already over.
The word is well attested: Proto-Germanic *sahsa-, Old High German sahs, Old English seax, Old Norse sax. “Scramasax”, by contrast, appears exactly once in an ancient source: Gregory of Tours, Historia Francorum IV,51 (cultris validis, quos vulgo scramasaxos vocant), plus a derived passage in the Liber Historiae Francorum 35. The type series Kurzsax – Schmalsax I/II – light and heavy Breitsax – Langsax with the measurements given is standard scholarship (Böhner 1958; Hübener 1988; Wernard 1998; Siegmund 1998; summarised by Westphal, RGA 26, 2004). Siegmund's Lower Rhine chronology comprises eleven phases between the mid-fifth and mid-eighth centuries, based on 316 male and 215 female graves. Krefeld-Gellep has over 6,000 excavated graves from the first to the eighth century. The Seax of Beagnoth (British Museum 1857,0623.1) measures 72.1 cm with a 55.1 cm blade at 985 g and carries the only complete epigraphic attestation of the twenty-eight character futhorc. The blade from Bad Krozingen (550–650) is a selectively hardened composite of core and jacket steel.
“The seax was the typical Viking weapon.” No. Its main period is the sixth to eighth centuries on the Continent, plus Anglo-Saxon England. For the Viking Age proper, four long knives are known from Denmark, around twenty from Norway and eighteen from Gotland — against hundreds of swords. “The Saxons are named after the seax.” That is Widukind of Corvey's story from the tenth century (Res gestae Saxonicae I,6–7), roughly half a millennium after the events; but the tribal name Saxones is already attested in Ptolemy in the second century, long before the Merovingian seax. A connection remains conceivable (Genrich; Rech), the direction of derivation is unresolved — Widukind's tale is folk etymology, not proof. “Scramasax is the technical term.” It is a single attestation with an unresolved word meaning; scholarship says long seax. Reenactment forms without evidence: the broken back is an Anglo-Saxon feature — Scandinavian long knives do not have it. Carrying was at the front or side, not on the back. Scabbards usually enclosed the grip as well, which is why a ring with a leather strap sat at the end of the handle; the bare handle of untreated antler is modern stage kit. And the old view of a crude, badly forged chopping knife has been refuted metallographically.
The seax is the most honest weapon of the Migration and Merovingian periods, because it never pretended to be anything grander. It grew out of the knife, stayed related to the knife, and still became the standard blade of half of Europe for centuries. It is at the same time the most useful find archaeology has: a clock made of iron that tells us when a person on the Lower Rhine was laid in the ground.
And it carries the earliest sentences craftsmen wrote about their own work — “Biorhtelm made me”, a rune row with typos, a name called Beagnoth about whom nobody knows anything more precise to this day. Anyone fond of blades and letters will find in these pieces the origin of an idea that has not changed since: whoever makes something well wants their name on it.
Weapon Lore · The Viking Sword · Viking Smithing · Warfare · The Saxons · The Franks · The Spear.
All History & Finds articles →
Few images from the Viking world stir as much excitement – and as much argument – as the shield-maiden: the woman with …
Read more →The Viking battle was no wild scramble but a hard struggle shield to shield. A look at tactics and weapons beyond the …
Read more →A great deal has been written about Viking swords: about pattern welding, about inlaid blade inscriptions, about hilt …
Read more →