by Yuri Kuchinsky


Greetings, friends,

Here's how Matthew 4:1 reads according to our standard canonical Greek text.

(RSV Mt 4:1) Then Jesus was led up by the Spirit into the wilderness to be tempted by the devil.
In contrast, this is the Aramaic version of this passage, according to the Curetonian manuscript.
(Mt 4:1 Curetonian) Then Jesus was led by the Holy Spirit into the wilderness to be tempted by Satan.
Thus, we can see that, according to this ancient Aramaic text, it was the "Holy Spirit" that was leading Jesus into the wilderness.

Well, I happen to think that this is one clear case where Aramaic version is more original than the Greek. For one thing, it's certainly a lot easier to see why the word "Holy" would be omitted from this verse by a later editor, than why it would have been added later. After all, as the early Christian movement kept growing by leaps and bounds, and becoming more theologically sophisticated, some authorities may well have objected to the idea that the Holy Spirit wanted Jesus to be tempted by the devil. And, on the other hand, it's not at all easy to see why anyone would have wanted to add this word at a later date -- had it not been already there in the original version of Matthew. Within the context of Christianity rapidly becoming more mainstream within the Roman empire, the theological significance of the Temptation scene could hardly have been increasing -- rather the opposite seems more likely.

And there are also many other arguments that can be made to the same effect. But in order to analyse this problem in some depth, we will also need to consider what the other Synoptic gospels say in their parallel passages. And besides that, we will need to examine some additional textual evidence -- the evidence from our earliest manuscripts and versions. We will soon see that it's not only the Aramaic Matthew that happens to feature the word "Holy" in this passage, but numerous other versions as well.

So, to begin with, let us see how this passage reads in all three canonical Greek gospels. Some important differences among them are underlined below.

(RSV Matthew 4:1) Then Jesus was led up by the Spirit into the wilderness to be tempted by the devil.
(RSV Mark 1:12) The Spirit immediately drove him out into the wilderness. (13) And he was in the wilderness forty days, tempted by Satan;
(RSV Luke 4:1) And Jesus, full of the Holy Spirit, returned from the Jordan, and was led by the Spirit (2) for forty days in the wilderness, tempted by the devil.
As we can see, all three canonical gospels lack the word "Holy" in this passage. But there's also another problem that is now emerging in this Synoptic comparison, and it's quite serious for our mainstream 2 Source Theory (the hypothesis that, independently of each other, both Mt and Lk were originally based on Mk). Because it can be observed that both Mt and Lk use the passive voice here when referring to the action of the Spirit, whereas Mk uses the active voice.

ACTIVE VOICE: the Spirit drove him out (Mk)
PASSIVE VOICE: he was led by the Spirit (Mt, Lk)

In Greek:

ACTIVE VOICE: to pneuma auton ekballei (Mk)
PASSIVE VOICE: anecthe upo tou pneumatos (Mt, Lk)

And so, in Mark, "the Spirit" is the subject of the sentence, and Jesus is the object. And in the other two Synoptics, Jesus is the subject, and "the Spirit" is part of a passive clause. But, on the 2 Source Theory, this wasn't supposed to happen!

Unlike our first problem with "the Spirit" vs. "the Holy Spirit", in this case there's no perceptible difference in meaning, or any great theological significance in choosing one expression over the other; essentially, these are just two ways to say the same thing. But, upon examining a wide range of early textual evidence, it turns out that our early NT gospels tradition is pretty well split along these two lines in treating this passage, with many important Western (Syro-Latin) witnesses featuring an active voice construction.

And so, in the present study, these two problems will now be considered together, with the view of determining what was the earliest version of this passage in the earliest Christian gospel. Yes, we can now even go beyond what the earliest version of Matthew said -- and go straight to the earliest source of Matthew. There now appears to be enough evidence overall even for this. (In any case, most scholars today believe that Matthew wasn't really the earliest gospel, so this part shouldn't really be too controversial.)

Let us begin by dealing with our second problem first, i.e. the question of the active vs. passive voice in regard to the Spirit. After all, no complex theological questions need to be dealt with in this connection, so this can simplify our analysis quite a bit. Whatever subsequent editorial decisions would have been made in this case, they seem to have been of a purely grammatical nature, and thus somewhat easier to identify.


So, at first glance, this looks like one of those famous "anti-Markan agreements", since both Mt and Lk happen to agree here with each other over against Mk in placing the "Spirit" as part of a passive clause. And this will be quite difficult to explain based on our dominant 2 Source Theory -- to which something like 90% of all biblical scholars currently subscribe. According to the 2ST, both Mt and Lk depended originally on Mk, so they shouldn't really agree among themselves against Mk at all!

(Well, generally speaking, such logical arguments against the 2ST don't really vex our mainstream biblical establishment too much. After all, there are something like 1000 of such anti-Markan agreements of Mt and Lk -- also known as "the Minor Agreements"! While, in a normal world, this, in itself, would have already been quite sufficient to bury 2ST, in the world of biblical scholarship, on the other hand, it's not really seen as anything that one would worry about too much... A citation from the evangelist seems to be quite appropriate here, "Blessed are they who did not see, and yet believed...")

Thus, if we suspended all pre-conceived theoretical judgements in this case, the evidence before us seems to indicate that the confluence of Mt and Lk is preserving for us here a more original version of this passage. On this reasoning, Mark's version would be the latest of the three.

I'd say that, in general, such an argument would be pretty hard to go against. But -- while, indeed, it will probably be valid for the overwhelming majority of these anti-Markan agreements -- upon a closer analysis, this case appears to be different after all... So, instead, I will now argue that, despite these seeming odds, the Greek Mk in this case does actually represent the earliest version of this passage over against Mt and Lk!

What led me to this conclusion is primarily the testimony of the ancient Aramaic gospels, the Curetonian and the Sinaitic. So here they are, in Burkitt's literal translation. (The Curetonian version survives only for Mt in this case.)

(Mt 4:1 OS Curetonian) Then Jesus was led by the Holy Spirit into the wilderness, to be tempted by Satan.
(Mt 4:1 OS Sinaitic) Then Jesus was led by the Spirit into the wilderness, to be tempted by the Accuser (akelqartsa).

(Mk 1:12 OS Sinaitic) the Spirit sent him forth into the wilderness; and he was there forty days being tempted by Satan.

(Lk 4:1 OS Sinaitic) Now Jesus, being filled with the Holy Spirit returned from the Jordan; and the Holy Spirit took him and sent him forth into the wilderness, that he might be tempted by Satan.

The most important item of evidence here, from my perspective, is the testimony of the Old Syriac Aramaic Luke, which happens to be quite different from the canonical Greek Luke. As we can see, Aramaic Luke agrees with Mark (both Greek and Aramaic Mk) rather than with Matthew in this reading that we are now investigating (active vs. passive voice). Thus, as a result, the majority reading among the OS Aramaic texts favours the active voice when describing the action of the Spirit, whereas the majority Greek reading goes otherwise.

So this leads me to the conclusion that in the earliest version of this sentence, "the Spirit" was probably the subject of the sentence.

And, of course, OS Aramaic Luke also happens to feature the word "Holy" when referring to the Spirit that led Jesus into the wilderness. (More about this later on.)


Below, you can see a Table where 4 main variant readings for this passage are all brought together, as they are attested by a variety of gospel manuscripts.

For example, in the first horizontal line, we will find that the following manuscripts feature "the Holy Spirit".

In Matthew: Old Syriac, Hebrew, and the Peshitta versions.
In Mark: the Greek Codex Bezae.
In Luke: Old Syriac.

Furthermore, the following Diatessaronic versions also feature this phrase: the Magdalene Gospel, Ephrem the Syrian, the Persian Diatessaron, and the Dutch Diatessaron.


Mt Mk Lk MG Eph Per Dut Aph Arb
1. Holy Spirit AHP  Bez  A + + + +    
2. Spirit G AGP GP         + +
1. Active v. for "Spirit"   AGP APB + + +   + +
2. Passive voice AHGP   G       +    

A = Aramaic gospels (Old Syriac)
H = Hebrew Matthew (Howard's text)
G = canonical Greek text
P = Peshitta
Bez = Greek Codex Bezae
B = Bohairic Coptic version
MG = Magdalene Gospel
Eph = Ephrem the Syrian
Per = Persian Diatessaron
Dut = Dutch Diatessaron
Aph = Aphrahat
Arb = Arabic Diatessaron

The second horizontal line of the Table represents those manuscripts that read simply "the Spirit" in this passage.

The third horizontal line represents those manuscripts where active voice is used to describe the action of the Spirit

The fourth line represents those where the Spirit forms part of a passive clause (i.e. where Jesus is the subject of the sentence).


Clearly, for our two case studies, the majority of the Old Syriac, and of the early Diatessaronic textual tradition does favour the word "Holy" for this passage, as well as the active voice for "Spirit" (these are the readings that are each numbered as #1). And there are also other valuable witnesses that happen to support these Old Syriac and Diatessaronic readings. Thus, I would argue that they represent the earliest version of this passage.

Ephrem the Syrian, and Aphrahat are both very important early Syriac commentators, and their testimony is highly respected by textual scholars. And also the Persian Diatessaron is generally taken to represent the Syriac textual tradition. Thus, we have quite a large number of agreements among different representatives of Semitic textual tradition, which indicates that these two readings marked as #1 are very early indeed.

Many people still assume that the Diatessaron originated only around 170 CE. But there's now an increasing awareness that gospel harmonies already existed well before that, so there's no good reason to think that the date ca. 170 CE should have any big significance in this regard. (It seems like, broadly speaking, 170 CE should really represent the terminus ante quem, rather than post quem for dating the earliest gospels harmony.)

Also, it can be naturally assumed that the earliest harmony of the gospels was based on, and incorporated within itself some very primitive gospel texts -- the texts that are otherwise no longer extant. Thus, in my view, the sort of a text that is represented by the confluence of Ephrem and of the early Western Diatessarons will in fact often represent the earliest pre-canonical text of the gospels.

So, in my view, it is the original gospel text that is really at stake in this investigation, rather than some mythical "Tatian's Diatessaron".


Here's how this passage is found in a variety of Diatessaronic texts.

So led him the Holy Spirit that he was [away] from the people in the desert, for [him] to be tempted by the devil.

EPHREM THE SYRIAN (according to Burkitt; this is based on the Armenian text, as the original Syriac is not extant for this passage):
Immediately the Holy Spirit took and led him out into a desert, to be tempted by Satan

PERSIAN DT (G. Messina edition):
Then the Holy Spirit carried Jesus to the desert, so that the Devil might tempt him.
(Allora lo Sancto Spiritu porto Gesu nel deserto, affinche il diavolo lo tentasse)

LIEGE DT (Plooij edition):
When Jesus had been baptised, he was led by the Holy Ghost into the wilderness to be tempted by the evil spirit.

APHRAHAT, IN SYRIAC (according to Burkitt):
And then the Spirit sent him forth that he might be tempted by Satan

(4:42) And Jesus returned from the Jordan, filled with the Holy Spirit. (43) And immediately the Spirit took him out into the wilderness, to be tried of the devil;

Thus, the earliest version of this passage seems to have included both of the features that we still find in our OS Aramaic Luke: it was the Holy Spirit that was leading Jesus into the wilderness, and the Holy Spirit was the subject of the sentence.

We may suppose that, thereupon, two basic changes were made in this passage -- and they seem to have been made in different circumstances, for different reasons, and by different individuals.

-- At some very early point in time, the editors (or perhaps even the authors?) of Matthew decided to make Jesus the subject of this sentence instead. While the reason for this isn't really entirely clear, still, it didn't really change the meaning of the sentence in any big way. So this seems to have been a purely stylistic change; it dominates the entire Matthean textual tradition, hence it must have been very early indeed. And this reading is even reflected in the Dutch Diatessaron (but not in any other Diatessaronic version among the 6 in our sample!).

And yet, "Spirit" used as the subject of the sentence -- which I see as the earlier version -- is still found in the Greek Mk, and this same usage still happens to dominate the majority of the Semitic textual tradition, besides also being found in the Bohairic Lk. The change in the Greek Lk to agree with Mt in this passage seems to have been the latest of them all.

-- In regard to "the Holy Spirit", the change to simply "the Spirit" seems to have been theologically motivated. Here we see the OS textual tradition and the Diatessarons still overwhelmingly supporting "the Holy Spirit" -- while the Peshitta reads with the Greek in both Mk and Lk, along with the OS Mk. (So this would be just one of many cases where the Peshitta had been assimilated to the canonical Greek texts.)

One may suspect that this particular alteration in the Markan textual tradition was very early indeed, since it's already attested for us in the Aramaic Mk. And yet, the testimony of the Greek Codex Bezae -- the word "Holy" in its version of Mk 1:12 -- is quite significant. So it looks like Bezae still preserves the original text here. And the same can be said about the testimony of the Hebrew Matthew.

The removal of the word "Holy" from the Greek Luke seems to have been done as the latest of all these changes -- so this harmonised Greek Luke with the other two Greek Synoptics in this respect.

Furthermore, we have here the following Diatessaronic testimony for "the Holy Spirit",

Magdalene Gospel
Ephrem the Syrian
Persian Diatessaron
Dutch Diatessaron

So this pretty well establishes that this was how this passage read in the earliest gospel harmony.

It would be highly unlikely that the ancient editors of all of these witnesses had the canonical text before them, and then they all happened to make such a change independently of each other. It's far more likely, on the other hand, that this Western text reading represents the original gospel text.


So "the Holy Spirit" seems to have stood in this passage originally. And the reasons for removing the word "Holy" are not really all that difficult to surmise. I would date that change to ca. 200 CE. As Christianity was rapidly becoming ever more mainstream and "respectable" within the Roman empire, so was no doubt also the theological status of Jesus the Christ, the founder of the movement. Yeshua, once the lowly Galilean preacher, was now rapidly moving towards the exalted status of the Pantocrator -- the reigning King of the Universe -- which he did eventually acquire under the early Byzantine empire (the Council of Chalcedon, held in 451 CE, seems to have been especially influential in this regard).

One suspects that the Temptation scene, itself -- full as it is of the Tanakh parallels -- may have been perceived as somewhat embarrassing by the mainstream Catholic exegetes of those later times. Why would the Holy Spirit want Christ Jesus -- now seen as the Pre-existent Creator of the World, the glorious King of the Universe -- to be tempted by the devil? And this right after the baptism by some fellow named John, which also seemed to impart plenty of importance to the baptism scene, in and of itself? (And yet we do know that the baptism of Jesus by John was perceived as somewhat embarrassing by at least some of the later canonical editors -- hence its absence in both Luke and John.)

So I think that such theological reservations on the part of some Christian authorities may have provided sufficient incentive for reducing "the Holy Spirit" in this passage to simply "the Spirit".

And how would one explain the reverse process, i.e. why would anyone want to add this word at a later date? Well, even if some such reason might be conjectured after all, still, the much more difficult task would then be to explain how could such an idea have occurred to so many editors in different parts of the world, and working in all these different languages to boot -- who all happened to be editing Western texts?

Thus, inevitably, in order to make such an alternative hypothesis respectable -- in order to give it at least some historical verisimilitude -- one would have to postulate that all these cases of "the Holy Spirit" do depend on some sort of a shared source, after all... This is what the principle of parsimony seems to demand here.

So this shared source was obviously very early, since our Diatessarons do attest it already at ca. 150 CE. And, in order to account for a wide geographical distribution of this variant reading, this source would have had to be pretty influential. I conclude that the simplest explanation here would be that this mysterious early source was ultimately none other than the original source of the Synoptic gospels, itself -- the original Synoptic proto-gospel.

And the second variant reading that we have been considering here, "Spirit" as the subject of this sentence, likewise probably originally belonged to the same very early source.

Best regards,

Yuri Kuchinsky.

Click here to return to Yuri's Ancient Aramaic Texts page.