Does Arami Oved Avi refer to a wandering / poor Aramean, or to Lavan who sought to destroy?

This post deals with an interpretation of Arami Oved Avi by Ibn Ezra and Radak which goes against the classic midrashic interpretation, and the reaction of two supercommentaries of Rashi to this "daring" interpretation. What comes into play is whether Ibn Ezra and Radak can claim to have absolute knowledge of Hebrew to be able to declare the midrashic interpretation to not work out according to the rules of dikduk; and whether one can argue on midrash, as they are doing, if after the midrashic interpretation goes all the way back to Sinai! It could also be that as supercommentators of Rashi, they are simply defending Rashi's interpretation as one of peshat.

At the start of Ki Tavo, we hear of Arami Oved Avi:

ה וְעָנִיתָ וְאָמַרְתָּ לִפְנֵי ה' אֱלֹהֶיךָ, אֲרַמִּי אֹבֵד אָבִי, וַיֵּרֶד מִצְרַיְמָה, וַיָּגָר שָׁם בִּמְתֵי מְעָט; וַיְהִי-שָׁם, לְגוֹי גָּדוֹל עָצוּם וָרָב.

5 And thou shalt speak and say before the LORD thy God: 'A wandering Aramean was my father, and he went down into Egypt, and sojourned there, few in number; and he became there a nation, great, mighty, and populous.

Rashi cites the famous midrash:

An Aramean [sought to] destroy my forefather: [The declarer] mentions [here] the kind deeds of the Omnipresent [by stating]:“An Aramean [sought to] destroy my forefather.” That is, Laban, when he pursued Jacob, sought to uproot [i.e., annihilate] all [the Jews], and since he intended to do so, the Omnipresent considered it as though he had actually done it (Sifrei 26:5), for [regarding] the pagan nations of the world, the Holy One, Blessed is He, considers the [mere] intention [of an evil deed] as [being equivalent to] the actual perpetration [of the deed itself]. — [Yerushalmi Pe’ah 1:1 at end]

ארמי אבד אבי: מזכיר חסדי המקום ארמי אובד אבי, לבן בקש לעקור את הכל, כשרדף אחר יעקב. ובשביל שחשב לעשות, חשב לו המקום כאלו עשה, שאומות העולם חושב להם הקב"ה מחשבה [רעה] כמעשה:

The Sifrei indeed says something like this:

ואמרת לפני ה׳ אלהיך ארמי אובד אבי • מלמד שלא ירד יעקב לארם אלא

ב)(לאובד) • ומעלה על לבן הארמי כאלו איבדו)

where Gra emends to להאבד. Though I would assert that there is room for movement here. But the simple reading is that אובד here is being taken as a transitive verb. ("Bill grew tomatoes" is transitive, while "Bill grew" is intransitive.) And so it is being taken as a statement that "An Aramean destroyed, or caused to wander, my father Yaakov.

Ibn Ezra takes exception to this. He writes:

אובד אבי -

מלת אובד מהפעלים שאינם יוצאים ואילו היה ארמי על לבן היה הכתוב אומר: מאביד או מאבד.

ועוד: מה טעם לאמר לבן בקש להאביד אבי וירד מצרימה ולבן לא סבב לרדת אל מצרים, והקרוב: שארמי הוא יעקב, כאלו אמר הכתוב: כאשר היה אבי בארם היה אובד.

והטעם: עני בלא ממון.

וכן: תנו שכר לאובד והעד: ישתה וישכח רישו.

והנה הוא ארמי אובד היה אבי.

והטעם: כי לא ירשתי הארץ מאבי כי עני היה כאשר בא אל ארם, גם גר היה במצרים והוא היה במתי מעט, ואחר כן שב לגוי גדול ואתה ה' הוצאתנו מעבדות ותתן לנו ארץ טובה.

ואל יטעון טוען: איך יקרא יעקב ארמי?

והנה כמוהו יתרא הישמעאלי והוא ישראלי, כי כן כתוב.

Radak as well endorses this position. Thus:

והוא בודד כשהוא

מן הקל, כי פירוש פסוק ארמי אבד אבי יעקב אבי היה

אובד וקראו ארמי, לפי שגר בארם ושם היה אובד כי בצער

גדול היה שם עשרים שנה כמו שהתרעם על לבן ואמר הייתי ביום

אכלני חרב וקרח בלילה (בראשית ל א , מ . ) וכאשר

הפעל הזה מן הכבדים הוא עובר

That is, there is a difference between the kal and kaved patterns of this verb. The kal is intransitive, while the kaved is the transitive. And not only does this make sense grammatically, but other things in context or elsewhere work to demonstrate that this is the peshat. Thus Radak shows how it manifests in the actual story with Yaakov and Lavan, and Ibn shows how it fits in with the general theme being developed in the mikra bikkurim that we started out small with nothing. So dikduk plus working out with message make for good peshat.

As I noted elsewhere, Rashi's peshat, which is based on Sifrei, works well with the trup. Because the trup, which is Pashta Munach Zakef-Katon creates a situation of:

Arami | Oved Avi

as opposed to Mahpach Pashta Zakef Katon, which would be

Arami Oved | Avi

Of course, that trup could have been written to be in line with the midrash, rather than the peshat.

I would agree with Ibn Ezra and Radak as to the peshat meaning. And I would not discard Rashi's explanation as wrong. Rather, it is midrash, rather than peshat. And that is fine. Midrash need not always work out in accordance with the strict dictates of dikduk.

Mizrachi and Gur Aryeh both strive to defend Rashi. I am not convinced that they succeeded, or even that a defense based on dikduk is needed.

But Mizrachi writes about this, noting the Sifrei, as well as Ibn Ezra and Radak. His motivation is that since this is a midrash, from Chazal, this was transmitted via tradition, one man to the next, all the way back to Moshe Rabbenu who heard it from the Almighty, that the Arami is Lavan and than the Oved here is a transitive rather than intransitive verb, such that Lavan sought to destroy my father.

Of course, I would interject that there are other weighty opinions that do not consider every midrash to be of this sort, as a tradition all the way back to Har Sinai.

Mizrachi seeks to undermine Radak's grammatical point by pointing out that in Radak's own introduction, he writes that our knowledge of the Hebrew language is incomplete and reconstructed from the 24 books of Tanach, from Mishnaic Hebrew, and the like. So there could be a meaning which we would not know from other Biblical evidence. And so oved can be a transitive verb, and our unbroken midrashic tradition can be the true meaning of the pasuk, even on a peshat level.

This is of course true. Anything is possible. See for a practical example the kal passive, and how people arrogantly claim something is impossible and outside the bounds of Hebrew grammar, without focusing on this possibility. Still, we have other forms, namely kaved, to get where we want, and no evidence of this particular meaning within this form. And it is not just that, but bringing in Lavan out of left field does not seem very peshat-like, while Ibn Ezra's explanation works quite well in terms of the peshat.

Mizrachi also points to a pasuk towards the end of Ki Savo, in perek 28, which reads:

כ יְשַׁלַּח ה' בְּךָ אֶת-הַמְּאֵרָה אֶת-הַמְּהוּמָה, וְאֶת-הַמִּגְעֶרֶת, בְּכָל-מִשְׁלַח יָדְךָ, אֲשֶׁר תַּעֲשֶׂה--עַד הִשָּׁמֶדְךָ וְעַד-אֲבָדְךָ מַהֵר, מִפְּנֵי רֹעַ מַעֲלָלֶיךָ אֲשֶׁר עֲזַבְתָּנִי.

20 The LORD will send upon thee cursing, discomfiture, and rebuke, in all that thou puttest thy hand unto to do, until thou be destroyed, and until thou perish quickly; because of the evil of thy doings, whereby thou hast forsaken Me.

which he interprets as until he is אבד you. This is not how JPS chooses to render it; instead, it is "until you perish quickly." Indeed, the verb pattern of hishamedcha which precedes it implies a passive, which then takes an actor out of the picture. Mizrachi seems to explicitly rule out that possibility.

On the other hand, compare with other instances of שמד and אבד in this perek. In particular:

מח וְעָבַדְתָּ אֶת-אֹיְבֶיךָ, אֲשֶׁר יְשַׁלְּחֶנּוּ יְהוָה בָּךְ, בְּרָעָב וּבְצָמָא וּבְעֵירֹם, וּבְחֹסֶר כֹּל; וְנָתַן עֹל בַּרְזֶל, עַל-צַוָּארֶךָ, עַד הִשְׁמִידוֹ, אֹתָךְ.

48 therefore shalt thou serve thine enemy whom the LORD shall send against thee, in hunger, and in thirst, and in nakedness, and in want of all things; and he shall put a yoke of iron upon thy neck, until he have destroyed thee.

where there is an actor; as well as:

סג וְהָיָה כַּאֲשֶׁר-שָׂשׂ יְהוָה עֲלֵיכֶם, לְהֵיטִיב אֶתְכֶם וּלְהַרְבּוֹת אֶתְכֶם--כֵּן יָשִׂישׂ יְהוָה עֲלֵיכֶם, לְהַאֲבִיד אֶתְכֶם וּלְהַשְׁמִיד אֶתְכֶם; וְנִסַּחְתֶּם מֵעַל הָאֲדָמָה, אֲשֶׁר-אַתָּה בָא-שָׁמָּה לְרִשְׁתָּהּ.

63 And it shall come to pass, that as the LORD rejoiced over you to do you good, and to multiply you; so the LORD will rejoice over you to cause you to perish, and to destroy you; and ye shall be plucked from off the land whither thou goest in to possess it.

But these are different verb patterns, and I am not convinced that this pasuk Mizrachi cited is evidence that אבד in the kal pattern is a transitive verb.

Mizrachi also argues (see inside) with Ibn Ezra's assertion that Lavan trying to destroy my father is out of place and unconnected to context. (I would interject that midrashically, Tg Yonatan elsewhere understands that Lavan went down to Egypt under the name of Bilaam, which handily explains how this all fits in. But Rashi, and Mizrachi, are not considering this possibility.) Who says that there needs to be a connection, other than that it is a chesed that Hashem did? Though in the end, he sort of finds a way to work it into the context and flow.

Even so, I agree with the thematic point Ibn Ezra is making, and that if one reads it in Ibn Ezra's way, it advances the theme (of rags to riches) much more, while with Rashi and Sifrei it does not.

The Maharal mi-Prague, in Gur Aryeh, also comes to Rashi's defense and attacks Ibn Ezra and Radak. He leads off explaining just how it does work out in context -- that it is a separate detailing of the chesed that Hashem did. This in answer to Ibn Ezra's left-field comment.

Meanwhile, Ibn Ezra wants to differ from the chachmei haEmet, and say that it should have said מאבד אבי and that as written, it means that he was without anything. But the Maharal does not think very much of Ibn Ezra's proof, but that one can ask some very good questions on Ibn Ezra.

One can answer Ibn Ezra that אובד is indeed not a transitive verb, but is a noun. Because Lavan always thought how to destroy my father, he is the destroyer of my father. Thus, it is not that an Aramean destroyed my father, but that an Aramean was a destroyer of my father. But if it said מאבד, it would imply that it was a one time effort. And this is on the pattern of {Bemidbar 24:20}:

כ וַיַּרְא, אֶת-עֲמָלֵק, וַיִּשָּׂא מְשָׁלוֹ, וַיֹּאמַר: רֵאשִׁית גּוֹיִם עֲמָלֵק, וְאַחֲרִיתוֹ עֲדֵי אֹבֵד.

20 And he looked on Amalek, and took up his parable, and said: Amalek was the first of the nations; but his end shall come to destruction.

which Radak wrote in the name of the scholar Rabbi Yaakov ben Eliezer; and its meaning is that he always thought to destroy him, and therefore he is called the אובד of his. For in many places the harm caused by some thing is called by the name of the thing. And a stone, where it rests in a place where it can cause people to stumble is called a michshol, not a machshil. So too here, he is called the oved of my father because he is the אבידת יעקב. And so too in many places the actor of some thing is called by the name of the thing.

Furthermore, baalei dikduk often write about words that it is a strange word. So why should they be bothered here if they think it does not work out according to their dikduk? And he gives an example of a word -- ויסר -- needing to be explained as transitive though it is usually instransitive. {See Ibn Ezra's explanation of ויסר there, in his commentary on Shemot 14; and what he writes in his commentary of Bereishit 8; Imy impression is that Ibn Ezra is not saying that this is a kal which is usually intransitive but suddenly yotzei, but that it is from the binyan hakaved hanosaf and is strange because of a vowel substitution.}

Further, there are other examples of אובד which seems to be transitive, even though Ibn Ezra forces an intransitive explanation. Thus, in Devarim 32:28:

כח כִּי-גוֹי אֹבַד עֵצוֹת, הֵמָּה; {ר} וְאֵין בָּהֶם, תְּבוּנָה. {ס}

28 For they are a nation void of counsel, and there is no understanding in them.

which the Maharal feels should be a transitive verb, even though Ibn Ezra forces himself to explain otherwise, writing:

כי גוי אובד עצות -
הנכון שזה הגוי הם צרי ישראל.

ומלת אבד -
שם התאר והיא סמוכה, על כן נפתח הבי"ת וידוע כי מלת אבד מהפעלים העומדים, רק פירוש עצות שב אל אבד, כמו: נובלת עליה.
וכן: ותאבדו דרך, שהוא סמוך אל דרך והנה השכיל המתרגם שאמר: מאבדי עצה.

making it into an intransitive; but ultimately agreeing with the metargem who renders it transitive.

I would disagree with Maharal here and say that Ibn Ezra's explanation there is compelling rather than forced; and that the metargem had to switch it to a different mishkal in order to make it transitive, but this was cleverness in making the translated text easier.

And, he continues, Ibn Ezra and Radak should have explained the word in this way and should not have differed from the words of the Chachamim and the Targumist.

And he continues with further arguments how it can grammatically work out -- he only sought to destroy my father, but did not succeed, so it should stay as an intransitive.

Then, he turns to context -- what cause is there for mentioning Lavan's attempts to destroy my father? In answer: Lavan wanted to entirely destroy my father entirely, but Hashem saved him, and he descended to Egypt. And that it does not mention Hashem saving him from the hand of the Aramean as it mentions Hashem redeeming them from the hand of Egypt, this is because it does not want to make two good things. The point is that he was in these various dire straits and now he was given a wonderful land. It is starting with genai and ending with shevach.

Thus, he concludes, the words of Chazal are correct. And if it is difficult for you to believe that Lavan was trying to destroy Yaakov, despite chasing after him (because this was not necessarily to destroy him), see

the sefer Gevurat Hashem on the Hagaddah shel Pesach, and there (see here, I suppose, though I don't see it there) you can see it explained with clear proofs.

However, Ibn Ezra's explanation has neither hands nor feet, for he wrote that Yaakov is called an Aramean, and why should he be called an Aramean, where Avraham was called Avraham the Ivri -- even though he was in Eretz Yisrael, he was called by the initial name. So why does it call Yaakov an Aramean, and why does it not say "In Aram, my father was אובד?" And this question is greater than the question that he asks on our Rabbis, za"l, that it should have written מאבד. And furthermore, in which place do we find that the meaning of Oved means pauper? For the proofs which he brings from תנו שכר לאובד is no proof.

{This based on Mishlei 31:6:

תְּנוּ-שֵׁכָר לְאוֹבֵד; וְיַיִן, לְמָרֵי נָפֶשׁ.	6 Give strong drink unto him that is ready to perish, and wine unto the bitter in soul;
ז יִשְׁתֶּה, וְיִשְׁכַּח רִישׁוֹ; וַעֲמָלוֹ, לֹא יִזְכָּר-עוֹד.	7 Let him drink, and forget his poverty, and remember his misery no more. {P}

for what is written after it that "he shall forget his רישו, and his effort he will remember no more," the explanation of this is not lack of money, but rather the explanation of "he shall forget his רישו" is all that he lacks. And behold, anyone who lacks something is called a ריש; all the more so when he is lacking more. But for someone who lacks money to be called an אובד, this is simply not so!

And further, this that Ibn Ezra writes {in explaining the import of Arami Oved Avi}, "for I have not inherited the land from my fathers; for behold when Yaakov was in the land he was lacking, without clothing {?}", this matter has neither {good} taste nor smell, for even though when he was in Aram he was lacking money, it was possible that he would have the money when he returned from Padan Aram, for he was a very wealthy man. And what of it that he was lacking in Aram?

So ends my rough summary of the points of Mizrachi and Maharal mi-Prague. Now, my brief thoughts on the matter.

Firstly, I am somewhat biased towards Radak and Ibn Ezra. Seeing them attacked for what seem to be reasons of frumkeit, that they are not entitled to hold their positions because of pious concerns, sparks that reaction in me. It also makes me tend to view those who attack Ibn Ezra and Radak's position as likely to be biased, and therefore not really evaluating the merits of either side appropriately. Despite all this, I do think that Radak and Ibn Ezra are correct, more or less.

Let me be frank. I think that Radak and Ibn Ezra would laugh at many of the suggestions raised by Mizrachi and Maharal. Radak and Ibn Ezra are real grammarians, and the suggestions being raised are farfetched and implausible. It seems somewhat possible to me that continued attempts to explain Rashi's concerns as peshat, where Rashi passes midrash off as peshat, makes one view midrashic concerns as real peshat / dikduk concerns.

It may be that there is some way to reconcile the midrash with a grammatical reading of the verse. But the midrash saw Arami, connected it to the famous Lavan haArami, and worked from there. But this is not the most straightforward reading of the verse, that one would arrive at without the midrash. Neither Maharal nor Mizrachi would likely be endorsing this explanation if the Sifrei, and thus what they consider the tradition miSinai, were at stake.

We have no real Biblical evidence that אובד works as a transitive verb, and we have transitive versions which are used, though not used in this instance. And Lavan trying to destroy Yaakov, as opposed to e.g. Esav trying to destroy Yaakov, or Yosef being sold, or Dinah being captured, or Avraham fighting against the kings, or many other chasadim, seems random. I don't like random. There is no cause for it in context. Meanwhile, to say that it describes my forefather's sorry state before Hashem's salvation works out wonderfully in the context.

Radak and Ibn Ezra are not being driven by trying to preserve midrashic tradition. That is not what the process of peshat commentary is about. It is about looking at the text, mustering the tools of dikduk and context, and deciding what the text most probably means. A reference to Lavan is not likely, so it is not true that Radak and Ibn Ezra should have found some way to work it out.

And many of the suggestions they posit are quite far-fetched and irregular. That does not recommend a peshat commentary. Thus, Mizrachi suggests a pasuk later in Ki Tavo which is אובד in transitive. But this is quite close context, and one would think that expert grammarians such as Ibn Ezra and Radak would have picked up this close example. Yet they didn't, and they should not have, for we see from context that it was likely not intended as a transitive verb. And Mizrachi's suggestion that Radak should have humbly recognized what he said in his own introduction, that knowledge of the Hebrew language is not complete -- sure, anything is possible, and for all we know Bereishit Bara Elokim Et Hashamayim veEt HaAretz could really translate to "In the middle, God sipped coffee and tea." And each of these words are simply unattested to in these meanings attested to in the 24 books of Tanach we possess. So anything is theoretically possible. Should this be driving a pashtan? No, for if so, he would not be able to say anything about any verse; or else he could say anything about any verse. The existence of a midrash should not be a driving factor, if midrash need not be peshat.

What about Maharal? He suggests that it is a noun, rather than a transitive verb. And that a noun can be called based on what it causes, rather than as the causer. And it still works out perfectly with the trup, I would add! Also theoretically possible, and a very creative end-run around the rules of dikduk. As Ibn Ezra notes elsewhere, there is no such thing as a present-tense verb in Biblical Hebrew. There is the neutral-tense verb, which is really simply the noun. Ani Shomer means "I am watching..." or else "I am a watchman." The noun is the neutral tense verb. And based on context, that neutral tense can also be the present tense or future tense. So of course one can claim that it is the noun rather than the verb. But by doing so, one can make an end-run around any troublesome verb, and make it mean what you want it do mean. There would not be -- at least in the neutral tense -- transitive or instransitive, or causitive or passive. With enough creativity, it can mean whatever you want it to mean. Just declare it a noun!

But reading the pasuk, it does not really seem to be functioning as a noun. Forget the creative reinterpretation which makes it unlikely, the awkwardness of the construction makes it unlikely. Should a pashtan seize on it as the peshat meaning of the pasuk? Not unless he is biased towards justifying the midrashic reading, and he is ideally not biased in this way. (They would argue that he should be biased, since it is "possible" plus we have a Sinaitic tradition. But not every midrash is mipi haGevurah, some say; and even if it is, not every midrash is peshat.)

The idea that Lavan didn't actually carry it out, so it should be in the intransitive form is justified, is creative; but creative is not necessarily true or plausible. Do we have any other examples of this phenomenon? This is a midrashist playing with dikduk and calling it peshat. I have a similar reaction to the claim that it was decided that it be a noun because of continuous effort, rather than a one-time effort. This is creative, but I don't think it would be convincing to pashtan grammarians such as Ibn Ezra and Radak, and for good reason. Though we do have certain other forms (such as imperfect) to denote continuous action.

In terms of the questions on Ibn Ezra based on meaning, some seem to me to be good, and some not so good. Even the "good" ones are not as convincing to me as they appear to be to the Maharal. Because sometimes textual interpretations have difficulties, but they are still on the whole better than the alternatives. Despite Yaakov being able to buy the land on his return, once he was no longer a pauper, the point and theme was to stress how he was wandering about, and was not settled yet in the land. Why he would be called an Arami does not really trouble me; many objections can be raised, and are raised, on minor debateable points by midrashically inclined commentators. That does not make them solid peshat concerns. And even where there are solid peshat concerns, those can often be the result of the fine-grained peshat decisions of the particular commentator, but another explanation can be put forth which obviates those solid concerns and not resorting to farfetched midrash and dikduk.

Indeed, I would agree with Shadal, somewhat, who said:

ה ] אבי : כולל כל האבות כאחד שהיו תועים מגוי אל גוי , והראשון בא מארם , וקרובין לזה דברי הרשב " ם

and with Rashbam, who said:

ארמי אובד אבי -

אבי אברהם ארמי היה, אובד וגולה מארץ ארם. כדכתיב: לך לך מארצך.

וכדכתיב ויהי כאשר התעו אותי אלהים מבית אבי - לשון אובד ותועה אחד הם באדם הגולה, כדכתיב: תעיתי כשה אובד בקש עבדך.

צאן אובדות היו עמי. רועיהם התעום, כלומר מארץ נכריה באו אבותינו לארץ הזאת ונתנה הקב"ה לנו.

Where Maharal focused his attack on the idea of it being Yaakov who was the Arami Oved, he did not focus on Rashbam, who agrees but makes it Avraham. And he did not anticipate Shadal, who came later.

Indeed, historically the patriarchal wanderings may have come at about the same time as general Aramean migrations. And Avraham came from Aram. He was indeed a wandering Aramean. And it could mean Avraham, Yitchak, Yaakov, and perhaps even the shevatim. They came initially from Aram, and are called Arameans since they did not assume the name of their destination, as they never permanently settled there. So my forefathers were wandering Arameans. And there wanderings took them to Egypt. And so on and so forth. We can make them paupers, or wanderers, and answer some of Maharal's questions. And eventually they were redeemed from Egypt and because plentiful and inherited the entire land.

It works out perfectly, or near perfectly. I say Ibn Ezra, Radak, Rashbam, and Shadal win. Of course, Mizrachi and Maharal cluttered the page with impressive seeming arguments, and those who don't have an ear for dikduk and peshat may well be convinced of the opposite position. I still say that Ibn Ezra, Radak et al have not been toppled.