either monogenetic or polygenetic, separative error, which separates A, B, and
D further from C.62 All three cases indicate that manuscript C adheres more
closely to the Latin text.
In chapter XXI.2, after Satan is cast out of Hell and the patriarchs sing the
Tollite portas verses derived from Psalm 24(23):7–963 in chorus, exhorting Hell
to open its large gates for the arrival of Christ, King David recalls to the patri-
archs that when he was alive on earth, he pronounced the words commem-
orating the power of the Redeemer derived from the well-known Psalm
107(106):15–16. The reading of C 28r/26, “er ek var lifs ꜳ iordu” (“when I was
alive on earth”), is certainly the closest variant to that of the Latin text, which
in this place reads T99r/7–8 “Nonne cum essem uiuus in terris predixi uobis”
(“When I was alive on earth, did I not foretell you”), whereas the reading
shared by A, B, and D, “þa er ec lifða” (“when I lived”), seems to transmit a
considerable trivialization of the original reading.
A 53v/27–8
B 11v/25–6
D 6v/22
C 28ra/26
T 99r/7–8
þa er ec lifða64
þa er ec lifþa65
þá er eg lifda66
er ek var lifs ꜳ
cum essem uiuus
iordu67
in terris68
In chapter XXVI, after the patriarchs have been freed from the bondage of
sin and have been delivered to Paradise, they see the Good Thief Dismas walk-
ing towards them and bearing a cross on his shoulders. The lectio difficilior is
again transmitted in C, where the present participle C 28vb/15 “hafandi” (“hav-
ing, carrying”) reflects the present participle of the Latin text T 103r/18 “por-
tans” (“carrying”). On the other hand, A, B, and D share a considerably less
learned relative clause, “sa hafði” (“who had, carried”), formed by the demon-
strative pronoun “sá” and the praeteritum of the verb “hafa.”
A 55r/27
B 14r/13
D 9r/9
C 28vb/15
T 103r/18
sa hafði69
sa hafði70
hann hafde71
hafandi72
portans73
In chapter XXVI, after the patriarchs’ encounter with Dismas, A, B, and D
omit a clause of the text which survives in C 28v/15–16, “Þa er Guds helgir
sꜳ þenna man spurdu þeir” (“But when the saints of God saw that man, they
The Manuscript Filiation of Niðrstigningar saga 45
asked”), translating the Latin T, 103r/11–12 “Uidentes autem sancti omnes
dixerunt ad eum” (“Moreover, when all the saints saw [him], they told him”).
A 55r/28
B 14r/14
D 9r/10
C 28vb/15–16
T 103r/11–12
om.
om.
om.
Þa er Guds
Uidentes autem
helgir sꜳ þenna
sancti omnes
man spurdu
dixerunt ad
þeir74
eum75
Possible Significant Errors of A, C, and D
Manuscripts A, C, and D share three possible conjunctive and separative er-
rors76 and a single separative error77 against the readings of B, whose text seems
in all cases to be more accurate.78
Nevertheless, the accuracy of B cannot be conclusive evidence in determin-
ing its authority, as the first two readings belong to the fourth interpolation and
are consequently absent in the Latin source text T.79 However, they are also
absent in the texts of Revelation 12:9 and Augustine’s Sermo 265D, from
where the interpolation ultimately derives.80 As a matter of fact, it is more like-
ly that the first three conjunctive errors presented here are in fact the original
readings transmitted in the common ancestor of A, B, C, and D, and that the
diverging readings of B are the result of a subsequent correction of its text,
possibly ex ingenio, as suggested below.
In chapter XXI.1, within the fourth and last interpolation of the Icelandic
text, which relates to the transformation of Satan into a terrifying dragon and
to his travel to Jerusalem, where Christ had just been crucified, B 11v/12 re-
ports that Satan “flo hann þangat” (“and he flew there”), whereas A, C, and D
have “oc for þangat” (“and [he] travelled there”). Since Satan had just acquired
the shape of a great dragon, it is plausible that the scribe of B may have de-
cided to correct the original generic verb “for” (“travelled”) with “flo” (“flew”),
specifying that Satan’s travel from Hell to Jerusalem was a flight rather than
simple journey.
A 53v/14
C 28ra/11
D 6v/3–4
B 11v/12
T 101r
oc for þangat81
ok for þangat82
fór Satan
oc flo hann
–
þangad83
þangat84
Again, in the above-mentioned interpolation of chapter XXI.1, manuscripts
A, C, and D may transmit a possible trivialization of the text. Their narratives
describe how at the sight of Christ’s dead body on the cross, Satan wanted to
46 Niðrstigningar saga
“tear away the soul” of Christ (“slita ondina”), whereas B 11v/12–13 reads “to
steal the soul” (“slęgia ǫndina”). The verb “slægia,” literally meaning “to clean
out fish,” found in B (here meaning “to steal”) may be an embellishment of the
verb “slita” (“to tear away”) shared by the rest of the tradition. The latter verb
may have been perceived as either erroneous or inappropriate with reference to
the soul of Christ. Within this connection, it should be noted that the fourth
interpolation was probably inspired by Augustine’s Sermo 265D, which relates
to the Devil being “greedy and avid for the death” of Christ (“Mortis avidus
diabolus fuit, mortis avarus diabolus fuit”), and that rather than aiming to steal
the soul of Christ, Satan was craving to “swallow” it (“devorare”/“gløpa”) like
a voracious animal.85 Hence, the more dramatic verb “slita” (“to tear away”)
may have been the one present in the common ancestor of A, B, C, and D,
which was subsequently revised in B with the considerably less incisive verb
“slægia” (“to steal”).
A 53v/14–15
C 28ra/12
D 6v/5
B 11v/12–13
T 101r
slita ondina
slita mundu
slíta ỏndina frá
slęgia aundina
–
þegar fra
aundina fra
hỏnum88
fra honum89
honom86
Ihesu87
In the above-mentioned chapter XXI, when the patriarchs encounter the
Good Thief in Paradise carrying his heavy cross, B 14r/13 reads “otirligr maþr”
(“wretched man”), which precisely renders the Latin T 103r/11 “vir miserri-
mus” (“most wretched/miserable man”). This adjective was employed in the
Latin text to describe his physical appearance, as he is later asked by the patri-
archs T 103r/12, “Quis es tu anime quia uisio tua latronis est?” (
“Whoever are
you, O soul, for your face is that of a thief?”). A, C, and D, on the other hand,
seem to transmit a common error, “maþr allosęlligr” (“most joyless/ill-favored
man”), that does not fit the context and could possibly have been an attempt of
the first redactor to calque the Latin “miserrimus.” As a matter of fact, at the
moment of narration, Dismas is neither sorrowful nor ill-favoured, since he
was granted entrance to Paradise before any other mortal soul (except for
Enoch and Elijah), and being in Paradise, he has already reached a state of
bliss. Nevertheless, “otirligr” (“inglorious”) could possibly be a second inter-
vention of the scribe of B, possibly unsatisfied with the less elegant adjective
“allosæligr” (“most joyless”) shared by the rest of the tradition.
A 55r/27
C 28vb/14–15
D 9r/9
B 14r/13
T 103r/11
maþr
maþr
madr allsiálligr92
otirligr maþr93
vir miserrimus94
allosęlligr90
allosælligr91
The Manuscript Filiation of Niðrstigningar saga 47
Although unstable in all three manuscripts, the reading below can be count-
ed as a separative error of A, C, and D against B. In chapter XXV, the patriarchs
encounter Enoch and Elijah, who had been already living in Paradise, and in-
form them of the prophecy foretelling their future death at the hands of the
Antichrist. A, C, and D transmit a secondary reading, which specifies that be-
fore fighting the Antichrist, they shall first come down to earth from Paradise:
A 55r/22–3 “Enn þa munom viþ ðanga⟨t⟩ coma” (“And then we shall come
there”). Being a separative error, this omission may have aroused indepen-
dently from that of the Latin source text T.
A 55r/22–3
C 28vb/10
D 9r/3
B 14r/9
T 103r/8
Enn þa munom
Enn þa skulum
Vid fỏrum þa til
om.
om.
viþ ðanga⟨t⟩
vid fara nídr i
ydar97
coma95
heiminn96
Possible Significant Errors of A and B
There is a single instance in which A may share with B a monogenetic, con-
junctive, and separative error against D.
In chapter XXIV.3, Habakkuk recalls one of his canticles in praise of God,
as related in Habakkuk 3:13: “existi in salutem populi tui ad liberandos electos
tuos” (“you came out for the salvation of your people and set free your chosen
ones”).98 The text of D in this case transmits the correct reading, “útvalda þina”
(“your chosen ones”), using the past participle of the verb “útvelja” (“to choose
/ to select”) reflecting the Latin “electos” (“chosen ones”). A and B, on the
other hand, transmit a significantly different reading “valaþa ðina” (“your poor
ones”), which differs considerably from the source, T 102v/21 “ad liberandos
electos tuos” (“to set free your chosen ones”). However, being a section of text
extracted from a renowned canticle, this reading may have been easily emend-
ed ex ingenio in D (or in its antigraph) during the time of their transcriptions.
A 55r/8
B 13v/14
D 8v/9
C
T 102v/21
at leysa valaþa
at leysa valaþa
ad leisa útvalda
lacuna
ad liberandos
ðina99
þina100
þina101
electos tuos
Possible Significant Errors of A and C
A single monogenetic, conjunctive, and separative error is shared by A and C
against B in the Latin text of Psalm 107(106):15–16, transmitted in chapter
XXI.2. The third person plural subjunctive of the original text, T 101r/8–9
48 Niðrstigningar saga
“Confiteantur Domino” (“They would confess to the Lord”), is well preserved
in B but corrupted into a second person plural, indicative as “Confitemini
Domino” (“You confess to the Lord”), in both A and C. This reading, as in the
previous case, being a section of a well-known psalm, may have been easily
emended by the scribe of B during the time of its transcription.
A 53v/29
C 28r/27
B 12r/1
D 6v/23
T 101r/8–9
Confitemini
Confitemini
Confiteantur
om.
Confiteantur
Domino102
Domino103
Domino104
Domino105
Errors of A and D
In a single case, A and D share a monogenetic, conjunctive, and separative er-
ror,106 which further highlights their filiation from a common ancestor and
separates them from B and C. Moreover, there are two separative errors that
separate them from B.107
In chapter XXVI, the Latin text relates that the Good Thief was accompa-
nied in Paradise by the archangel Michael, who exhorts him to wait there for
the arrival of the patriarchs, who will shortly thereafter be freed from Hell.
Michael pronounces the words “modicum sustine” (“wait a while”), which are
still well preserved in both B and C: “scalltu biþa litla stund” (“you shall wait
a while”). On the other hand, A and D seem to share a trivialization in substi-
tuting the second person plural “scalltu” (“you shall”), with the first person
plural of the modal “skulu” (“shall”): “scolom biþa litla stund” (“we [two]
shall wait a while”).
A 55v/9
D 9v/4–5
B 14v/5
C 28vb/27
T 103r/23
scolom biþa litla skulum vid bída
scalltu biþa litla
skaltu nu bida
modicum
litla stund109
stund110
litla stund111
sustine112
In chapter XVIII.1, A and D omit one of the five invectives pronounced by
Hell against Satan, T 102r/1 “sputum iustorum” (“spittle of the just”), which is
still preserved in B, where it is rendered more metaphorically as “hrøptr af
monnom” (“defamed by men”).113
A 54v/5
D 7v/12
B 12v/21
C
T 102r/1
om.
om.
hrøptr af
lacuna
sputum
monnom114
iustorum115
The Manuscript Filiation of Niðrstigningar saga 49
In chapter XXIV.3, A and D omit the conclusive line of the prophecy voiced
by Micah after the deliverance of the patriarchs from Hell, T 102v/28 “sicut
iurasti patribus nostris” (“as you promised to our fathers”), derived from Micah
7:18–20. The clause is fully extant in B: “sva sem sem þu svaraþir feþrom
orom” (“as you promised to our fathers”).
A
D
B 13v/21–2
C
T 102v/28
om.
om.
sva sem sem þu
lacuna
sicut iurasti
svaraþir feþrom
patribus
orom116
nostris117
Stemmata Codicum
Turville-Petre’s Stemma
As mentioned above, Turville-Petre rightly believed that the four medieval
manuscripts A, B, C, and E derived from a common original but that E must
have been thoroughly emended and revised on the basis of the Latin text of
Evangelium Nicodemi. He also maintained that although the “first” and the
“second” interpolations (here referred to as the “third” and the “fourth”) de-
rived from the older redaction were intentionally maintained in E, the subse-
quent modifications of readings applied to E created a new textual facies
considerably different from that of A, B, and C. These considerations led him
to suspect contamination in the tradition of Niðrstigningar saga.118 Taking his
suggestions into account, his stemma could be similar to that proposed by Odd
Einar Haugen, as illustrated in the following section.
Aho’s Stemma
In his PhD dissertation, Gary L. Aho agrees with Turville-Petre in that the four
medieval manuscripts A, B, C, and E derive from a common original transla-
tion X, but he also argues that it is not necessary to postulate a contamination
in the tradition. He suggests instead that the text of X is more faithfully pre-
served in E, which stands alone in a single branch of the tradition, and con-
cludes that this may be the reason why its readings are so remarkably close to
the Latin text. He also maintains that A, B, C, and D all derive from a second-
ary, now lost, redaction of X (indicated as X1), which has further separated
them from the Latin text.119 His stemma is implemented in Figure 2.
50 Niðrstigningar saga
Latin A
X (First translation)
X (First redaction)
E
1
A
B
C
Figure 2. Gary L. Aho’s stemma
Haugen’s Stemma
Odd Einar Haugen, who has conducted the most thorough and comprehensive
analysis of the Niðrstigningar saga stemmatics, also recognizes that E should
be separated from A, B, C, and D and that it should be counted alone as a dif-
ferent translation. On the basis of the data discussed in his survey, Haugen
concludes that both Aho’s and Turville-Petre’s stemmata are plausible filia-
tions of the tradition of Niðrstigningar saga and that internal evidence is, in
this case, insufficient to reach any conclusive decisions. He then resorts to ex-
ternal evidence, such as the high presence of archaisms in both A and B, and
the dates of the four medieval manuscripts, which naturally indicate that the
revision was made on E rather than on A and B, since A and B are the oldest