Justin Martyr, who was martyred in the 160s, used this text in his composition Dialogue With Trypho, chapter 103. Commenting on Psalm 22, verse 14, he wrote, “In the memoirs which, I say, were drawn up by His apostles and those who followed them, it is recorded that His sweat fell down like drops of blood while He was praying.”
Reckoning that the Gospel of Luke was not written before the early 60s, this implies that Justin’s copy of the Gospel of Luke was separated from the autograph of the Gospel of Luke by less than a century.
About two decades after Justin, Irenaeus wrote the Third Book of his composition Against Heresies. In the 22nd chapter, Irenaeus used Luke 22:44, mentioning that if Jesus had taken nothing of Mary, that is, if He had not experienced a physical human nature, he would not have eaten food harvested from the earth, He would not have become hungry, or weary, “Nor would He have sweated great drops of blood.”
Irenaeus’ contemporary Tatian included Luke 22:43-44 in his Diatessaron, around the year 172. Around the year 360, as Ephrem Syrus composed his commentary on the Diatessaron, he also mentioned the detail about Jesus’ sweat becoming like drops of blood.
Also, in Ephrem’s Carmina Nisibena, in Hymn 35, part 18, Ephrem pictures the devil saying about Jesus, “While He was praying I saw Him and was glad, because He changed color and was afraid: His sweat was as drops of blood, because He felt that His day had come.”
In the early 200s, the writer Hippolytus referred to Luke 22:44, near the beginning of chapter 18 of Against Noetus. In the course of giving examples of the contrast between Jesus’ divinity and humanity, Hippolytus wrote that “In agony He sweats blood, and is strengthened by an angel.”
The first patristic writer to mention manuscripts that do not support Luke 22:43-44 is Hilary of Poitiers. Around 350, in Book 10 of his Latin composition De Trinitate, in part 41, Hilary wrote, “We cannot overlook that in very many Greek and Latin codices nothing is recorded about the angel’s coming, and the sweat like blood.”
Despite acknowledging such manuscripts, Hilary does not offer a judgment on whether the passage has been omitted in the copies where it is absent, or interpolated in the copies in which it is found. He seems to have been less concerned about reaching a correct verdict on the textual question and more concerned about promoting correct theology.
He said that heretics should not be encourage by the idea that Jesus’ weakness is confirmed by the need for an angel to strengthen Him, and that His sweat should not be construed as a sign of weakness. And like Irenaeus, he points out that the bloody sweat demonstrated the reality of Jesus’ physical body. When he states, “We are forced to the conclusion that all this happened on our account.” He seems content to use the text.
In 374, Epiphanius of Salamis made some very interesting statements about Luke 22:43-44. In Panarion 19:4, he quoted these verses an example of passages that Arians use to show that Jesus sometimes needed assistance from others, or that He was inferior to the Father: “And it says in the Gospel according to Luke, ‘There appeared an angel of the Lord strengthening Him when He was in agony, and He sweat; and His sweat was as it were drops of blood, when He went out to pray before His betrayal.”
It should be noticed that Epiphanius quoted verse 43 with the reading “angel of the Lord.”
In Panarion 61, Epiphanius used the passage again in the same way. He used the passage for doctrinal purposes, and stated that without the display of agony and sweat pouring from His body, the Manichaeans and Marcionites might seem reasonable in their theory that Christ was an apparition, and not completely real.” He emphasizes how Jesus’ sweat like blood showed that “His flesh was real, and not an apparition.”
Epiphanius claims in Panarion that Arius cited this very passage from the Gospel of Luke in an attempt to demonstrate the subordination of the Son to the Father.
So far, we could read Epiphanius’ remarks and think that the only form of the text he knew included verses 43 and 44. But in Ancoratus, chapter 31, Epiphanius wrote that the passage “is found in the Gospel according to Luke in unrevised copies.” Then he said, “The orthodox have removed the passage, frightened and not thinking about its significance.” Coming from someone who seemed ready to blame heretics for bad weather, this is a remarkable statement.
Epiphanius uses Luke 22:43-44 again in Ancoratus chapter 37 as evidence that Jesus was truly human, and that His sweat shows that He was physical.
Around the year 405 in
When this pagan writer objected to Jesus’ statement, “Do not fear those who kill the body,” he wrote that Jesus Himself, “being in agony,” prayed that His sufferings should pass from Him.” The term “being in agony” here is probably a recollection of Luke 22:43, because this term is used there, but not in the parallel-passages.
For the testimony of Amphilochius of Iconium, who lived from about 340 to about 400, we rely on a collection of extracts in the medieval manuscript Athous Vatopedi 507, from the 1100s. A note simply says: “Of Amphilochius bishop of Iconium, on the Gospel of Luke: it states there, “Being in agony, He prayed more earnestly.”
There is some reason to wonder whether Didymus the Blind, or someone else, was the author of the Greek composition called De Trinitate that is attributed him. Some interpretations of the author are different from interpretations expressed by Didymus in some other works. But, theologians do sometimes change their views. Whoever wrote De Trinitate, he made an accurate quotation of Luke 22:43 in Book 3, Part 21.
Ambrose of Milan, in the late 300s, in his commentary on Luke, seems to use a text that did not include verses 43-44; he does not mention the appearance of an angel and he does not mention that Jesus’ sweat became like drops of blood.
John Chrysostom is yet another patristic writer who used Luke 22:43-44. Once he did so in a comment on Psalm 109. And once he did so in the course of his 83rd Homily on the Gospel of Matthew, which covers the parallel-material in Matthew 26:36-38.
Homily 83 on Matthew, Chrysostom does
not say that he has put down the text of Matthew and has turned to the text of Luke. But after referring to Jesus’ prediction of
Peter’s denials, and Peter’s insistence that he will never deny Jesus,
Chrysostom transitions to the contents of Luke 22:43, stating, “And He prays
with earnestness, in order that the thing might not seem to be acting. And sweat flows over Him for the same cause
again, even that the heretics might not say this, that His agony was a
pretense. Therefore there is a sweat
like blood, and an angel appeared strengthening Him, and a thousand sure signs
After interpreting this for several sentences, Chrysostom returns to the text of Matthew 26:40.
We will reconsider the significance of this after we have seen the testimony of the cluster of manuscripts known as family 13.
For now, let’s go on to the next patristic reference.
The testimony of John Cassian should
not be overlooked, even though his name does not appear in the textual
apparatus for Luke 22:43-44 in the UBS Greek
New Testament or the Nestle-Aland compilation. John Cassian traveled widely: to the Holy Land, to
Jerome, in Against the Pelagians, Book 2, part 16, shows that he was aware of some copies that had Luke 22:43-44, and some copies that did not. In 383, he included this passage in the Vulgate. Later, in Against the Pelagians, he wrote that these words – the words we know as Luke 22:43-44 – are “In some copies, Greek as well as Latin, written by Luke,” which implies that Jerome also knew of copies in which the verses were not included.
Theodore of Mopsuestia, a
contemporary of Jerome who worked mainly in
Only slightly later comes Theodoret of Cyrrhus, who famously oversaw the withdraw of 200 copies of the Diatessaron in his churches. In 453, Theodoret wrote Haereticarum Fabularum Compendium, and in this work, after presenting Jesus’ statement in John 12:27, he says that Luke taught more clearly how Jesus was indeed suffering, when He was in agony, and he proceeds to use part of verse 44.
Now we come to the testimony of
Cyril of Alexandria, who died in the year 444.
In Cyril of Alexandria’s Sermon
146 and Sermon 147 on the Gospel
of Luke, Cyril describes the events in
He states, “Everywhere we find Jesus praying alone, you may also learn that we ought to talk with God over all with a quiet mind, and a heart calm and free from all disturbance.” This is not the sort of thing one says when one is reading a text that says that Jesus is praying in agony, and sweating huge drops of blood.
Cyril says in Sermon 147, “Let no man of understanding say that He offered these supplications as being in need of strength or help from another – for He is Himself the Father’s almighty strength and power.” Cyril does not come out and say that he rejects the idea that an angel appeared and strengthened Jesus, but he comes very close to doing so.
Severus of Antioch, in the first half of the 500s, supplies some additional information about the text used by Cyril. In an extract from the third letter of the sixth book that he wrote to “the glorious Caesaria,” Severus stated the following:
passage about the sweat and the drops of blood, know that in the divine and
evangelical Scriptures that are at
“‘But, since he said that the divine Luke inserted among his own words the statement that an angel stood and strengthened Jesus, and his sweat dripped like blood-drops or blood, let him learn from us that we have found nothing of this kind inserted in Luke’s work, unless perhaps an interpolation has been made from outside which is not genuine.
The books therefore that are among us contain nothing whatever of this kind. And so I consider it madness for us to say anything to him about these things. And it is a superfluous thing to oppose him regarding things that are not stated at all, and we shall be very justly condemned to be laughed at.’”
Then Severus says: “In the books therefore that are at
In this way, Severus
drew his reader’s attention to Emperor Julian’s use of the passage in the
mid-300s, and to Cyril of Alexandria’s rejection of the passage in the early
400s, and to the acceptance of the passage in
is particularly significant because he specifies that the copies in
Later, in the 600’s, a writer named Athanasius, Abbot of Sinai, is credited with yet another text-critically relevant statement about Luke 22:43-44. Amy Donaldson, in her 2009 dissertation, Explicit References to New Testament Variant Readings Among Greek and Latin Church Fathers, included his statement:
“Be aware that some attempted to delete the drops of blood, the sweat of Christ, from the Gospel of Luke and were not able. For those copies that lack the section are disproved by many and various gospels that have it; for in all the gospels of the nations it remains, and in most of the Greek.”
There is also a marginal note, preserved in minuscule 34, that states that “the report about the sweat-drops is not in some copies, but Dionysius the Areopagite, Gennadius of Constantinople, Epiphanius of Cyprus, and other holy fathers testify to it being in the text.”
We could examine more patristic support for Luke 22:43-44, from Augustine and Nestorius, for example. But let’s go back to the evidence from Chrysostom.
Why, in Homily 83 on Matthew, does he take a detour to comment on Luke 22:43-44? It cannot be absolutely ruled out that he just wanted to cover a parallel-passage. But another possibility is that by the time John Chrysostom wrote Homily 83 on Matthew, it was already customary that when the lector read the Gospels-reading for the Thursday of Holy Week, after reading Matthew 26:39, he also read Luke 22:43-44.
John’s brief detour into Luke 22 interlocks very snugly with this custom. In addition, in Codex C, a secondary hand has written the text of Luke 22:43-44 in the margin near Matthew 26:39.
This brings us to the evidence from
the cluster of manuscripts known as family 13.
In most members of family-13, Luke 22:43-44 appears in Luke, either in
the text or margin after Luke 22:42. Most
of the members of family 13 also have these two verses embedded in the text of
Matthew after 26:39.
The evidence from minuscule 1689, a member of family 13, is very helpful. This manuscript was lost for several years, but has been found safe and sound in the city of
Many other manuscripts have similar
notes in the margin at this point, as part of the lectionary apparatus.
It does not require a long leap to deduce what has happened in family 13: instead of resorting exclusively to margin-notes to instruct the lector to jump from Matthew 26:39 to Luke 22:43-44 and then return to Matthew 26:40, someone whose work influenced members of family 13 simplified things for the lector, by combining the parts of the lection in order within the text of Matthew.
Some commentaries have misrepresented this as if it implies that the passage is not genuine. But the evidence in family 13 just shows that a passage that was regarded as part of the text of Luke was embedded into the text of Matthew after 26:39 for liturgical purposes.
On a related point: when Luke 22:43-44 is accompanied by one or more asterisks, such as in minuscule 1216, the default deduction should not be that the purpose of the asterisks was to express scribal doubt, but to serve as part of the lectionary apparatus, drawing attention to the two verses that were to be read after Matthew 26:39 in the lection for Maundy Thursday.
So: was Luke 22:43-44 initially present, or initially absent? The passage is supported by a broad array of manuscripts, plus the manuscripts of over 20 patristic writers, and a couple of non-Christian writers. Four patristic writers – Hilary, Epiphanius, Jerome, and Athanasius of Sinai – show that they were aware that verses 43-44 were not supported in all copies, but nevertheless they favored the inclusion of the verses. Epiphanius even said that orthodox individuals had attempted to remove the passage.
One Latin writer – Ambrose of Milan – did not have verses 43 and 44 in his text of Luke 22.
And one Greek writer, Cyril of Alexandria, from the 400s, definitely did not have verses 43-44 in his text.
The most ancient evidence, from Justin, Tatian, and Irenaeus, includes the passage. The most geographically diverse support points in the same direction. And support for these verses does not come only from authors with only one doctrinal view.
Plus, internally, nothing in the
surrounding material calls for the insertion of additional material. Bart Ehrman has proposed that verses 43-44 do
not look like something Luke would write, on the grounds that Luke had an
interest in portraying Jesus as “imperturbable.” However, Luke reports about several actions of
Jesus in which His disposition is far from stoical or disinterested, including
His criticism of the synagogue-ruler in chapter 13, and His weeping over the
When we look at the external evidence that supports Luke 22:43-44, the question should not be “Did someone remove these verses from the text of Luke,” but “Why did someone remove these verses from the text of Luke?”
It is virtually unique to see a Christian writer assert that “the orthodox” tampered with the Gospels-text, and to imply that some orthodox believers revised the text in a way that was influenced by their fear.
In the 100s, the second-century writer Celsus, in a statement preserved by Origen, claimed that some believers “alter the original text of the gospel three or four or several times over, and they change its character to enable them to deny difficulties in face of criticism.”
There’s no way to tell if Celsus saw what he says he saw, but it can’t be ruled out that he did indeed notice Christians making changes to the Gospels-text, and that because some of those changes appeared to him to relieve perceived difficulties in the text, he naturally believed that this was the motivation for the changes.
However, he might have seen, and misunderstood, something else: textual adjustments that were not made to minimize interpretive difficulties, but to render the text easier to use when it was read in church-services.
One of those adjustments may have involved a liturgical feature pointed out by John Burgon in The Revision Revised. Here I slightly paraphrase his observations:
“In every known Greek Gospels lectionary, verses 43-44 of Luke 22 follow Matthew 26:39 in the reading for Maundy Thursday. In the same lectionaries, these verses are omitted from the reading for the Tuesday after Sexagesima – the Tuesday of the Cheese-eaters, as the those in the East call that day, when Luke 22:39-23:1 used to be read.
Furthermore, in all ancient copies of the Gospels which have been accommodated to ecclesiastical use, the reader of Luke 22 is invariably directed by a marginal note to skip over these two verses, and to proceed from verse 42 to verse 45.
What is more obvious, therefore, than that the removal of verses 43 and 44 from their proper place is explained as a side-effect of a lection-cycle of the early church?
Many manuscripts have been discovered since the time of Burgon, but in general, what he describes is accurate: Luke 22:43-44 is embedded after Matthew 26:39 in the lection for Maundy Thursday, and it is left out of the lection assigned to the Tuesday after Sexagesima Sunday.
The customary transfer of Luke 22:43-44 into the text of Matthew, when the text was read during Easter-week, may explain the sudden detour that Chrysostom took into this passage in the course of his Homily 83.
A scenario that explains the most evidence in the fewest steps is that when an attempt was made to revise the text for liturgical reading, one group of liturgical revisors took verses 43 and 44 out of Luke 22, but failed to re-insert them into Matthew 26. As soon as these verses dropped out of the text, the shorter reading was defended along the same lines that we see Cyril of Alexandria use to defend it.
We do not have hard evidence of this particular liturgical step of revision being undertaken in the second century, but the elegance of Burgon’s explanation is a strong factor in its favor. Plus, this theory accounts for the correspondence between this particular feature in the Easter-time lections, and the very similar contrast between forms of the text with and without the passage.
So: I conclude that Luke 22:43-44 was an original part of the Gospel of Luke.
I also conclude that its removal, in the second century, was probably not the result of some copyist’s desire to get rid of what he considered a problematic passage; nor was it the result of a heretic’s desire to remove a text that demonstrated the physicality of Jesus’ body. Instead, it occurred when orthodox believers transferred verses 43 and 44 into Matthew, after 26:39, conforming to their Easter-time custom, but failed to retain it in Luke, again reflecting their early Eastertime liturgy. As a result, these two verses fell out of the text.
This influenced texts known to Hilary, to Ambrose, and especially Cyril of Alexandria. It affected the text that was translated into Sahidic, and the Greek text that was translated into Armenian, and the Armenian text that was translated into Georgian. But as Athanasius the Abbot of Sinai stated, although some attempted to delete the drops of blood from the Gospel of Luke, the legitimacy of the passage is shown by the “many and various Gospels-manuscripts in which the passage is read.”
Luke 22:43-44 should therefore be respected and cherished for what it is: part of the Word of God.