If Njáls saga is the saga of law and Egils saga the saga of the poet, Laxdæla saga is the saga of feeling — the one with the most fully drawn women, the most interior emotional life, and the most famous closing line in all of Old Norse literature. Its great character is Guðrún Ósvífursdóttir, whose loves and griefs the saga traces across decades and whose final, devastating answer to her son is the sentence Icelanders still quote today. That makes Laxdæla the perfect text for two things English speakers find hardest in saga prose: the genitive machinery of descent and patronymics, which the saga lays on thickly in its genealogical openings, and the reported speech and reported thought that carry its introspective dialogues — and with them the subjunctive. We take the genealogical opening (chapter 1) and the deathbed dialogue (chapter 78). Read the orientation first if you have not (texts/saga-overview); the Njáls saga page drills the naming formula, so here we move quickly past it to descent and to reported speech.
Part 1 — The genealogical opening (chapter 1)
| Icelandic (normalised) | English |
|---|---|
| Ketill flatnefur hét maður, son Bjarnar bunu. | There was a man called Ketill Flat-Nose, the son of Björn Buna. |
| Hann var hersir ríkur í Noregi og kynstór. | He was a powerful chieftain in Norway and of great family. |
| Ketill flatnefur átti Yngvildi, dóttur Ketils veðurs, ágæts manns. | Ketill Flat-Nose was married to Yngvildur, the daughter of Ketill Wether, a man of distinction. |
| Unnur hin djúpúðga var enn dóttir Ketils, er átti Ólafur hvíti Ingjaldsson. | Unnur the Deep-Minded was likewise a daughter of Ketill's, whom Ólafur the White, Ingjaldsson, was married to. |
The opening formula is by now familiar — Ketill flatnefur hét maður "there was a man called Ketill Flat-Nose" (see the Njáls saga page for the formula itself). What Laxdæla does next, and keeps doing, is build a whole family in the genitive.
Descent and patronymics run on the genitive
A saga genealogy is a chain of genitives, and Laxdæla's first sentences are a clinic in it. Ketill flatnefur, son Bjarnar bunu — "the son of Björn Buna." The father's name Björn goes into the genitive Bjarnar (a u-stem genitive with the ö → ja shift), and the by-name buna into the genitive bunu. "Son of X" always puts X in the genitive; that is the engine of every Icelandic patronymic. You can see the patronymic in finished form in the same passage: Ingjaldsson "son of Ingjaldur" is literally Ingjalds (genitive) + son welded into a surname. (The genitive's full range: nouns/genitive-uses.)
Ketill flatnefur hét maður, son Bjarnar bunu.
There was a man called Ketill Flat-Nose, the son of Björn Buna. — son + genitive: Björn → Bjarnar (u-stem genitive), buna → bunu. 'Son of X' demands the genitive — the root of every patronymic.
Ketill flatnefur átti Yngvildi, dóttur Ketils veðurs, ágæts manns.
Ketill Flat-Nose married Yngvildi, the daughter of Ketill Wether, a man of distinction. — átti ('was married to') takes the accusative Yngvildi; then dóttur + the genitives Ketils veðurs ('of Ketill Wether'); ágæts manns is a genitive in apposition, 'of a distinguished man'.
Notice three case facts packed into the third sentence. átti ("had as wife / was married to") takes its spouse in the accusative: Yngvildi. Then dóttur ("daughter") governs the genitive of the father: Ketils veðurs ("of Ketill Wether"). And ágæts manns ("of a distinguished man") is a genitive in apposition to Ketils — adjective and noun both genitive, agreeing. This stacking of an accusative object with two layers of genitive descent inside one short sentence is the texture of saga genealogy, and it is precisely what overwhelms an English reader, because English marks none of it — "the daughter of Ketill Wether, a distinguished man" is all bare nominative-looking forms in English.
The free-standing article in by-names, and a relative er
Two more saga staples appear. Unnur hin djúpúðga — "Unnur the Deep-Minded" — uses the free-standing article hin (feminine) before the weak adjective djúpúðga, the standard epithet pattern ([Name] hinn/hin [adjective]; see register/literary-archaic). And the last clause hangs on **er: Unnur … er átti Ólafur hvíti — "Unnur … whom Ólafur the White was married to." Here er is the relative "who/whom" (= modern sem); the clause's own verb is átti, so er cannot be "is." The relative is doing real work: it relativises the object of átti, since átti takes its spouse in the accusative — Unnur is the one married, Ólafur the husband-subject. (The relative system: texts/saga-overview.)
Unnur hin djúpúðga var enn dóttir Ketils, er átti Ólafur hvíti Ingjaldsson.
Unnur the Deep-Minded was likewise a daughter of Ketill's, whom Ólafur the White, son of Ingjaldur, married. — hin djúpúðga = free-standing article + weak adjective ('the Deep-Minded'); er = relative 'whom' (clause verb átti); Ingjaldsson = Ingjalds (gen.) + son.
Part 2 — The deathbed dialogue (chapter 78)
Decades on, an old, now half-blind Guðrún sits with her son Bolli, who presses her to say which of her four husbands she loved most. The exchange, exactly as printed, builds to the most famous sentence in saga literature:
| Icelandic (normalised) | English |
|---|---|
| Þá mælti Bolli: "Muntu segja mér það, móðir, að mér er forvitni á að vita? Hverjum hefir þú manni mest unnt?" | Then Bolli said: "Will you tell me, mother, the thing I am curious to know? Which man have you loved most?" |
| Þá segir Bolli: "… hitt verður enn ekki sagt, hverjum þú unnir mest. Þarftu nú ekki að leyna því lengur." | Then Bolli says: "… but the other thing still has not been told — which man you loved most. You need not hide it any longer now." |
| Guðrún svarar: "Fast skorar þú þetta, sonur minn," segir Guðrún, "en ef eg skal það nokkurum segja, þá mun eg þig helst velja til þess." | Guðrún answers: "You press this hard, my son," says Guðrún, "but if I am to tell it to anyone, then it is you I will most choose for it." |
| Þá mælti Guðrún: "Þeim var eg verst, er eg unni mest." | Then Guðrún said: "To the one I loved most, I was worst." |
This is the introspective Laxdæla in full: reported question, hedged conditional answer, and a final aphorism whose grammar is as taut as its feeling.
Reported question and the dependent (reported) clause
Bolli's question is folded into a dependent clause: Muntu segja mér það … að mér er forvitni á að vita? — "Will you tell me the thing that I am curious to know?" And the buried question itself, Hverjum hefir þú manni mest unnt? ("which man have you loved most?"), then surfaces as an embedded interrogative in his next turn: hitt verður enn ekki sagt, hverjum þú unnir mest — "the other thing has not yet been told, which man you loved most." Two things to flag. First, the embedded verb unnir is subjunctive (the 2nd-singular subjunctive of unna "to love"), because it sits inside a reported/indirect question — the content is presented as what is to be told, not asserted as fact. This is the reported-speech subjunctive doing its perspective-marking work, which English cannot match (English keeps the indicative "which you loved most"). Second, unna "to love deeply" takes its beloved in the dative: hverjum ("to which [man]") is dative, not accusative. (Reported speech and its mood: complex/reported-speech and verbs/subjunctive-reported-speech.)
Muntu segja mér það, móðir, að mér er forvitni á að vita?
Will you tell me, mother, the thing I am curious to know? — muntu = munt ('will' 2sg) + þú; the relative-content clause að mér er forvitni á að vita ('that I have curiosity about knowing'); note mér er forvitni á, a dative-experiencer idiom.
… hitt verður enn ekki sagt, hverjum þú unnir mest.
… the other thing still has not been told — which man you loved most. — embedded reported question; the verb unnir is the 2sg SUBJUNCTIVE of unna (not indicative unnir), marking the indirect question; hverjum is DATIVE because unna ('love') governs the dative.
The conditional answer: subjunctive after ef
Guðrún hedges before she yields, and the hedge is built on a conditional. …*ef eg skal það nokkurum segja, þá mun eg þig helst velja til þess — "*if I am to tell it to anyone, then it is you I will most choose for it." The protasis with ef ("if") sets up a non-asserted, hypothetical condition; the apodosis þá mun eg… fronts þá ("then"), which throws the verb mun to second position (V2) ahead of the subject eg. The whole shape — ef P, þá Q — is the classic conditional skeleton, and the emotional weight sits in the reluctance the conditional encodes: she will tell, but only one person, and only this son.
Ef eg skal það nokkurum segja, þá mun eg þig helst velja til þess.
If I am to tell it to anyone, then it is you I will most choose for it. — ef-conditional protasis; the fronted þá ('then') in the apodosis triggers V2: verb mun before subject eg. nokkurum = dative 'to anyone'; velja takes accusative þig.
The aphorism: a fronted dative and a relative er
Now the famous line: Þeim var eg verst, er eg unni mest. Word for word: Þeim (dative, "to that one / to them") + var eg verst ("was I worst") + er eg unni mest ("whom I loved most"). The literal English is "To the one I loved most, I was worst," or more freely "I was worst to the one I loved most." Three pieces of grammar make it land.
First, the sentence opens with a fronted dative: Þeim ("to that one") is the indirect object/experiencer of vera verst ("be worst to"), pulled to the very front for emphasis. Because Icelandic is verb-second, that fronting throws the verb var ahead of the subject eg — Þeim *var eg verst — the same V2 inversion you meet everywhere (see Njáls saga's fronting). The dative þeim is the person *to whom she was worst; verst is the feminine superlative of vondur/illur "bad," agreeing with the speaker (Guðrún, feminine) as a predicate adjective.
Second, er is again the relative "whom": er eg unni mest — "whom I loved most." The clause's verb is unni (here the past tense / subjunctive of unna; in the famous line it reads as a simple past, "loved"), so er is relative, not "is." And once more unna governs the dative — the relativised "whom" is dative — which is exactly why the main clause can pick that referent up as the fronted dative þeim. The two datives (þeim in the main clause, the silent dative relativised by er) are the same person: the one she loved most is the one she was worst to.
Þeim var eg verst, er eg unni mest.
To the one I loved most, I was worst. — fronted dative Þeim ('to that one') triggers V2 (var before eg); verst = feminine superlative of vondur/illur agreeing with Guðrún; er = relative 'whom', and unna ('love') governs the dative, so 'the one I loved most' (þeim) is the same person she was worst to.
Honum var hún verst, er hún unni mest.
(parallel) To him she loved most, she was worst. — same template with honum (dative 'to him'); shows the construction is general: fronted dative experiencer + V2 + relative er + dative-governing unna.
The line is famous partly because of this grammar: the fronted dative makes the loved one the topic, the V2 inversion gives it the weight of a verdict, and the relative er binds love and cruelty into a single, inescapable clause. It is the saga's whole tragic logic compressed into seven words — and every one of those words is doing grammatical work English would need a paraphrase to carry.
Common Mistakes
❌ (reading) 'son Bjarnar bunu' — treating Bjarnar as a nominative name.
Case error — Bjarnar is the GENITIVE of Björn ('of Björn'), and bunu the genitive of the by-name buna. 'Son of X' takes the genitive; the nominative is Björn buna.
✅ (reading) 'son Bjarnar bunu' = 'the son of Björn Buna.'
Correct — son + genitive descent, the engine of every patronymic (Ingjaldsson = Ingjalds + son).
Saga genealogy is built in the genitive. Every "son/daughter of X" inflects X, and the chains stack — train your eye to read -ar / -s / -u descent endings as "of."
❌ (reading) 'hverjum þú unnir mest' — taking hverjum as accusative and unnir as indicative.
Two errors — hverjum is DATIVE (unna 'love' governs the dative), and unnir here is the SUBJUNCTIVE in a reported question, not a plain indicative.
✅ (reading) 'hverjum þú unnir mest' = '[reported] which man you loved most.'
Correct — dative hverjum + subjunctive unnir marking the indirect question.
The verb unna "to love deeply" takes the dative of the beloved, and inside a reported question its verb goes subjunctive — two things English (accusative-ish "whom," indicative "loved") never signals.
❌ (reading) 'Þeim var eg verst' = 'They were my worst' / 'Those were worst to me.'
Mis-parse — Þeim is the fronted DATIVE 'to that one', eg is the subject 'I', and verst agrees with eg: 'I was worst to that one'. The V2 inversion put var before eg.
✅ (reading) 'Þeim var eg verst, er eg unni mest' = 'I was worst to the one I loved most.'
Correct — fronted dative experiencer + V2 + relative er ('whom') with dative-governing unna.
Do not read the fronted dative þeim as the subject. The subject is eg ("I"); þeim is the dative "to that one," fronted for weight, which forces the verb in front of the subject.
❌ (reading) 'er eg unni mest' = 'and I love it most' / 'where I loved most.'
Mis-parse — er is the relative 'whom' (clause verb unni present); unna takes the dative, so the relativised 'whom' is dative, picking up þeim.
✅ (reading) 'er eg unni mest' = 'whom I loved most.'
Correct — relative er + dative-governing unna; the loved one is the same dative referent as þeim.
Key Takeaways
- Laxdæla saga is the saga of emotional and syntactic depth, built around Guðrún Ósvífursdóttir; its openings are dense genitive genealogy and its dialogues are dense reported speech.
- Descent and patronymics run on the genitive: son *Bjarnar bunu "son of Björn Buna," dóttur **Ketils veðurs" "daughter of Ketill Wether," Ingjalds*son" = "Ingjaldur's son." átti ("married") takes the accusative spouse.
- Reported and embedded questions take the subjunctive (hverjum þú *unnir mest "which you loved most"), and the verb *unna ("love") governs the dative (hverjum, þeim).
- The famous closing line Þeim var eg verst, er eg unni mest ("I was worst to the one I loved most") combines a fronted dative experiencer, the V2 inversion it forces (var eg), and the relative er ("whom") over a dative-governing verb — love and cruelty bound to one dative referent.
- Sources: Laxdæla saga ch. 1 and ch. 78, normalised text from the Icelandic Saga Database.
Now practice Icelandic
Reading grammar gets you part of the way. The exercises are where it sticks — free, no signup needed.
Start learning Icelandic→Related Topics
- Reading the Sagas: A Grammar GuideC1 — A practical cheat-sheet for reading Classical (Old/Norse) Icelandic saga prose, which modern Icelanders read with only modest help. Isolates the handful of grammatical features that differ from the modern language — the relative/temporal er (= sem/þegar), the historical present alternating with the preterite, the dense reported-speech subjunctive, the free-standing article hinn and bare nouns, the archaic and dual pronouns (vér/þér, vit/þit), and verb-initial narration with stylistic fronting. The headline: the sagas are grammatically close to modern Icelandic, so a B2/C1 learner can read them with this short list of switches.
- Long-Distance Reflexives: the Famous sigC1 — The construction that made Icelandic central to syntactic theory: the reflexive sig / sér / sín can be bound NON-LOCALLY — by the subject of a higher clause, across one or more clause boundaries — provided the intervening clauses are SUBJUNCTIVE. An indicative complement blocks the long-distance link and leaves only the local reading. The subjunctive, Icelandic's other flagship feature, is the licenser; one generalisation ties the two together.
- Reported Speech and Sequence of MoodB2 — The full machinery of indirect speech in Icelandic: the shift into the subjunctive, the backshift of tense into the PAST subjunctive under a past matrix verb, the adjustment of pronouns and deictics (hér to þar, í dag to þann dag, núna to þá), and reported questions (hvort / wh + subjunctive) and commands (að + subjunctive or infinitive). The key insight: Icelandic backshifts to the past SUBJUNCTIVE, not merely a past indicative as in English, so a single form væri encodes both pastness and reportedness.
- Annotated Saga: Njáls saga (Excerpt)C1 — A close grammatical reading of the famous opening of Brennu-Njáls saga — 'Mörður hét maður er kallaður var gígja' — in a normalised standard text from the Icelandic Saga Database. Annotates the predicate-first naming formula (X hét maður = 'there was a man called X'), the relative/temporal er, the free-standing article hinn in a by-name (Sighvats hins rauða), the terse paratactic style, and reported speech in the subjunctive, with an interlinear gloss line by line.
- Using the Genitive: Possession and BeyondB1 — What the genitive case DOES and where it sits in the sentence — the neutral postposed possessor (bók kennarans 'the teacher's book'), the partitive, governance by prepositions like til, án and vegna, and the meaningful contrast between the default postposed order and the emphatic preposed possessor (mín bók).
- Subjunctive in Reported SpeechB1 — The single most frequent subjunctive trigger in Icelandic: indirect speech introduced by að (and hvort/wh-words) after verbs of saying, thinking, hoping, and asking. The reported clause goes into the subjunctive to mark that the content is REPORTED, not asserted — present subjunctive (sé, komi, fari) under a present matrix verb, past subjunctive (væri, kæmi, færi) under a past one (backshift). Indicative can creep in for facts the speaker personally vouches for, making the mood a subtle evidentiality device.