Who Wrote the Late Volumes of Igrot Moshe?
Who Wrote the Late Volumes of Igrot Moshe?
By: Moshe Schorr[1]
Though this article deals with a factual question, it often seems to devolve into an ideological one. I therefore wish to state: I have no horse in this fight. I have not taken halakhic positions from Igrot Moshe volumes 7-9. I went into this with a genuinely open mind, and in the course of researching this question, I have taken the affirmative and negative sides of this question at different points.
Ever since Igrot Moshe volume 8 was published, and to a lesser degree volume 7, people have cast aspersions or directly accused it of being a forgery. The claim, generally, has been some variation of direct accusation or insinuation that somebody, usually either one of the Tendlers or R. Shabtai Rappaport, inserted his own teshuvot into the volume. Volume 9 is, as they say, ‘right out’. Some even call these volumes ‘Igrot Moshe David’.
As an example, Hirhurim several years ago published this quote from R. J.D. Bleich, though the comment thread is likely a better example.
Given the overwhelming consensus among latter-day authorities affirming the prohibition against drinking wine touched by a Sabbath-violator, Iggerot Moshe‘s position is surprising, to say the least. Moreover, the thesis developed in that responsum stands in sharp contradiction to Iggerot Moshe‘s earlier-cited multiple statements affirming the prohibition. Perplexed by Rabbi Feinstein’s surprising volte face, Rabbi Genut turned to a long-time, but unnamed, disciple of Rabbi Feinstein for clarification. Rabbi Genut quotes the disciple’s reply in which the latter writes that “it is known to me that many of the responsa [included in the posthumously-published eighth volume of Iggerot Moshe] were not before the eyes of my master and teacher… and there is also doubt with regard to many responsa in the seventh volume.”
The counterclaim, presented by the editors in the introduction to volume 8, is that the editors did exactly what their job entails: editing. While they added references, the teshuvot are by R. Moshe Feinstein.
I decided to test this. So the first thing I did was use an dataset given to me by Michael Pitkowsky, giving the dates, by year, of each teshuvah in volumes 1-8 of Igrot Moshe. This immediately yielded a stark result.
The spike in output in 1980-1981 is shocking. It is reminiscent of Barry Bonds’ late career.[2] It looks like a steroid year spike — how does a man in his eighties suddenly have more productivity than ever before? This, the first thing I saw, made me extremely suspicious. For comparison, Hatam Sofer’s chart looks like this:
I have published more on this at HaMapah, but suffice it to say: we expect to see a good deal of statistical noise in the amount of output,[3] though we do not expect to see changes that drastic, certainly not massive increases from authors in failing health.
This gave me the impetus to take the analysis a step further. So Avi Shmidman and I applied authorship analysis to it.[4]
Let me give a brief explanation of the algorithm. We are trying to look at the differentiability of the two classes. So, we take the 250 most common words, and then we look at the ability of a fairly standard model to separate the two classes. We expect to see some flukes or minor differences, so we’ll remove the most useful features — the words that are most predictive, and re-run. We will repeat this process ten times, removing three words each time. Different authors will have very substantially different linguistic usage — how often do you use the word ‘הוא’, ‘אבל’, etc., so even after removing the 30 most predictive words out of the 250 we’ll start with, it’ll be easily differentiated. However, with the same author, by this point the flukes should be gone, and we’ll lose any meaningful ability to differentiate the two classes. After the ten rounds we will be barely better than random guessing. (You’ll see at one point 58% accuracy — don’t be impressed — coin tossing is 50% accurate.)
Let’s start with our null hypothesis. Nobody, as far as I am aware, believes that the authors of Igrot Moshe and Minhat Yitzchak were one and the same. When running them against each other (IM vol. 6 vs Minhat Yitzchak), we get a final round accuracy of 97%. As we would expect. Now if we look at Igrot Moshe until we get to our “steroid spike”, if we compare the 60s and 70s to the 50s, we can get a sort of parallel null hypothesis.
So then we can just turn to our suspicious sets, and see where they fall.
Igrot Moshe volume 6, being the most recent undisputed volume, is the natural choice to benchmark here in terms of volumes. So let’s look at the three disputed volumes against volume six, and for good measure, let’s look at our “steroid spike” in 1980-1981 against the 60s and 70s.
The results are pretty clear. Bupkis. Nada. Zilch. None of the potential ways to slice and dice any of the potential forgeries turn up anything at all. And for the icing on the cake, most people who’ve learned Igrot Moshe would probably tell you that his prewar stuff is pretty different. Let’s compare the 20s against the 60s & 70s (the gray line).
So we see that not only are the differences between the new volumes and volume six minimal to the point of nonexistence, they’re far less differentiable than parts of his own corpus which are otherwise not under any suspicion are.
Let’s look at one last thing. Let’s look at our top ten features in favor of volume 9 over volume 6 when we tell them apart:
-
ע“א
-
ע“ב
-
ולכן
-
תמה
-
רש“י
-
בעניין
-
לו
-
התוספות
-
חייב
-
ד“ה
We generally consider ע”א, ע”ב, ד”ה — markers 1, 2, and 10 — to be markers of a good editor, and people pay good money for the expanded references in Mossad HaRav Kook editions. Numbers 5 and 8 are also components of references, as is 4, generally. So I’d like to suggest the following: the late volumes of Igrot Moshe bear substantial marks of editing. Having seen those, and generally getting a whiff of a difference, people justifiably viewed the late volumes of Igrot Moshe as tampered with, as fake even, with good reason, despite it just being editing. This isn’t without precedent. The common reyd, that the Terumat HaDeshen made up his own questions, has been disproven.[5] It seems to be from a similar reason – ‘good’ editing (as it was then considered) – stripping ‘unnecessary’ detail from the questions. So too here. The editor’s changes might be more immediately visible, but the consistent usage of simple function words – how often do you use function words like אני, הוא, זה, etc. — belies the true nature of the author.
Given
the preponderance of evidence that the later Igrot
Moshe volumes
are real (and spectacular), I think we can put the various theories
of alternative authorship to rest. The claims of the editors — that
the latest teshuvot were
dictated[6] — explains the ‘steroid spike’, and all available
evidence supports their central contention, that they didn’t change
the actual content. In short: it’s legit.
the preponderance of evidence that the later Igrot
Moshe volumes
are real (and spectacular), I think we can put the various theories
of alternative authorship to rest. The claims of the editors — that
the latest teshuvot were
dictated[6] — explains the ‘steroid spike’, and all available
evidence supports their central contention, that they didn’t change
the actual content. In short: it’s legit.
[1] Software
by Avi and Shaltiel Shmidman. Data from Michael Pitkowsky. Algorithm
as described in Koppel et al. (see below, footnote 4). With thanks to
Elli Fischer.
by Avi and Shaltiel Shmidman. Data from Michael Pitkowsky. Algorithm
as described in Koppel et al. (see below, footnote 4). With thanks to
Elli Fischer.
[2]
*
*
[3] To
clarify: I’m not saying it didn’t happen, just it’ll fluctuate
a lot without an actual cause or real reason.
clarify: I’m not saying it didn’t happen, just it’ll fluctuate
a lot without an actual cause or real reason.
[4] Koppel,
Schler, Bonchek-Dokow: “Measuring Differentiability: Unmasking
Pseudonymous Authors,” Journal of Machine Learning Research 8
(2007) 1261-1276.
Schler, Bonchek-Dokow: “Measuring Differentiability: Unmasking
Pseudonymous Authors,” Journal of Machine Learning Research 8
(2007) 1261-1276.
[5] J.
Freiman, Leket
Yosher,
Berlin ed. p.
XIV. http://hebrewbooks.org/pdfpager.aspx?req=8860&st=&pgnum=10
Freiman, Leket
Yosher,
Berlin ed. p.
XIV. http://hebrewbooks.org/pdfpager.aspx?req=8860&st=&pgnum=10
[6] See
volume 8, p. 3 in the introduction.
volume 8, p. 3 in the introduction.
32 thoughts on “Who Wrote the Late Volumes of Igrot Moshe?”
Can you compare it to Dibros Moshe as well?
A quick methodological query (and I admit that statistics and linguistic analysis are not at all my fields): Can this test discount a ‘good forgery’. In other words, if a different author made an effort to write in the same style as the original (or a close student who stuck closely to his master’s style) would it not give similar results as in the case the same author wrote both texts?
Is a disciple who worked with a master for decades and edits with permission someone creating a forgery? I think this question needs sharpness around definitions, and functionally definitions that materially matter. Especially since the modality for training and knowledge transmission relies on apprenticeship models and knowledge transmission that encourages the practices some would start to decry as forgery. Though, perhaps the calls of something being a forgery are motivated by something else (like disagreement or rejection of content, rather than facing the reality that it’s real).
The method is incredibly hard to fool. You need to know not the catchphrases — “lodestar” — but the regular usage of language, extremely precisely.
You must be mistaken. Much of IM 8 was written by the grandsons, as they write in the hakdama. The text that appears in smaller font was written by them to elucidate Rav Moshe’s words, which appear in larger font.
The fact that the method didn’t pick this up indicates that the method was in fact fooled, and the conclusion is therefore of no value.
The obvious counterargument is to assume (which is likely the case) that the data fed into the method from IM 8 was the text in larger type only.
I doub that that was the case, but without any clarification from the author, like I said, the conclusion is useless.
It’s also astonishing that the post makes no mention of this obvious issue, and does not tell us how they addressed it (which to me indicates that the author was not aware of it at all.)
Okay, I see there are some issues here.
A few things I should clear up, and I apologize for not making them clearer here.
1. The proportion of IM vol. 8 that is in small text is not actually that high. It’s substantial, but it’s nothing crazy. It also skews towards the stuff the classifier will disregard, because it’s often very specifically topical — like adding in block quotes from the sources, or filling in citations.
2. More fundamentally, generally speaking, the same works between authors display variance. There are a whole host of tokens that are very different between, say, Macbeth and Hamlet.
However, the topical ones — “Macbeth”, “Rosencrantz”, etc, are ones we wish to disregard here, because we need it to be robust across disparate documents by the same author. We want it to disregard minor changes, even if they are noticeable and recurring. An example: in the Ben Ish Hai’s Torah Lishma, every responsum ends in: “והיה זה שלום ואל שדי ה’ צבאות יעזור לי. כ”ד הקטן יחזקאל כחלי נר”ו. “.
The algorithm will disregard this blatant red flag and say, no, look at the overall usage of language, it’s really basically the same.
3. This one is perhaps the biggest omission, in my opinion, and I apologize for not being clear enough about this, the aim is not to prove that there aren’t editorial additions to volume 8. There are. My central point is the following. The much more serious accusation aimed at IM v. 8 is along the lines of what I quoted from R. Genuth — “it is known to me that many of the responsa [included in the posthumously-published eighth volume of Iggerot Moshe] were not before the eyes of my master and teacher” — or as a commenter on Hirhurim writes — “many of the later t’shuvos were ghost-written by the grandchildren based on what they thought RMF would hold.”
*This* is untrue.
This in no way disproves that there were tweaks, or tampering with individual words here and there, or even, for that matter, it wouldn’t rule out the existence of a handful of completely different teshuvot.
What it does disprove is the allegation that there are many teshuvot of the Tendlers inserted.
My apologies for the lack of clarity.
Your point is well taken, but I still think the methodology is flawed. It’s not just the small type. They also edited teshuvos, and some were apparently given to them baal peh to write up. If the style of those isn’t identified as different by the algorithm, either they wrote exactly in R. Moshe’s style, or the algorithm is not good. Either way, no proof can be deduced from it.
Where are the manuscripts from which these volumes were published? Comparing them to the printed edition would end all debate!
I got lost at “the differentiability of the two classes”.
What two classes?
“two classes” = the two text corpora which you are evaluating
As for the question on the large amount of output in later years, could not it have been that many responsa simply accrued over the years and had not been included in the previous volumes and now are? Admittedly, I have not checked the dates of the replies but it is a possibility.
As the post states, the study is based on the dates of the teshuvot, not when the volume appeared.
B”H
There were many substantial comments posted before the format of this blog was improved. They are no longer visible.
Hopefully they will be recovered.
Thanks.
Please go back to the old theme, now several comments are no longer available and some posts are with the layout completely misfit.
Fascinating. Perhaps to point out that the editors from the volumes of 5 and 6 are the same as the editors in the 7 and 8.
A comparison of the contents of volumes 1 and 2 compared with 7 and 8 may be more enlightening.
A few points I made in the previous blog comments version.
1. CS shu”t was also published posthumously.
2. The steroid spike” can be attributed to RMF’s A getting more inquiries as an older posek and even agudist daas Torah theories B he was now accepted as a leading posek in the 60s 70s 80s. C. He has more students than before (would make another potentially interesting statistical run.)
3. There was similar talk about vol 4 or 5 (cause it was printed in Israel?) so RMF printed in the beginning of the next volume a specific denial that it really was his work.
1. Indeed. It’s part of what the makes the RMF scandal a little unusual, since many, many shu”tim were published posthumously. Part of what makes this whole “you can’t trust posthumous shutim” angle strange.
2. No — it’s not mainly driven by number — he writes much longer answers.
RMF has a number of very long shu”tim. The long one on ribit from the 20s, over 50 double printed pages long I recall (side point, it seems everyone has a very long shut on ribit in early 20th century, obviously, the way to make your creds), in fact every shut on ribit is very long. And offhand, probably no one discusses ribit as much as RMF.
Anyway, statistically speaking, maybe the very long ones should be excluded from the steroid analysis. He does call them a choveret.
I wrote this comment previously, before the new switch to the new layout, and it has disappeared.
Something you don’t take into account is that it says in Igros Moshe 8 that much of it was written by the grandsons . if you look in the hakdama, they say that they inserted the text in small font in order to clarify what Rav Moshe wrote, which appears in the larger font. The fact that your algorithm shows that “Bupkis. Nada. Zilch.”, when we know and everyone admits that large portions of Igros Moshe 8 were written by others – that means that this method is faulty and/or not strong enough to detect a “good forgery”. These results therefore are of very litle value.
The design of this study is also flawed because no one is claiming that IM 8 and the ohter volues were written by different people, only that a small, undefined amount was written by others.
I also wrote a very lengthy comment about how this study should really be designed to overcome the “good forgery” issue, as well as other methodological issues.
If the blog owners could dig up that very lengthy comment, it would be appreciated.
This study is based on a faulty premise.
Nobody is claiming the entire sefer was completely forged, just that there are some tshuvos added in, that לאו גושפנקא דר’ משה חתים עלה.
For example, the Tshuva to Rabbi Eider on hilchos nidda is most definitely real, as is most probably any Tshuva addressed to a person still alive.
The entire idea is that some of the tshuvos addressed to the tendlers were added in.
The true test would be to compare specific thsuvos that people were מערער on with earlier volumes, (obviously with removing the small letters).
I believe with the Tshuva in question, the one about brain death, the entire thing is something like “I already told my son-in-law that we go with the Harvard criteria”, which would obviously be impossible to analyze.
You are correct that this study is fatally flawed. I wrote a lengthy comment under the previous layout that disappeared about the proper way to design this study.
I’m copying part of a clarification I made above:
My central point is the following. The much more serious accusation aimed at IM v. 8 is along the lines of what I quoted from R. Genuth — “it is known to me that many of the responsa [included in the posthumously-published eighth volume of Iggerot Moshe] were not before the eyes of my master and teacher” — or as a commenter on Hirhurim writes — “many of the later t’shuvos were ghost-written by the grandchildren based on what they thought RMF would hold.”
*This* is untrue.
This in no way disproves that there were tweaks, or tampering with individual words here and there, or even, for that matter, it wouldn’t rule out the existence of a handful of completely different teshuvot.
What it does disprove is the allegation that there are *many* teshuvot of the Tendlers inserted. There could certainly be a handful of fakes. But it’s not Igros Mordechai.
As for your specific test:
On an individual teshuva, it’s really not possible to very meaningful analysis. This can be done seriously on a decent number of them. But individual small pieces of writing are obviously more subject to randomness, and hence means of confirming them are less certain.
““it is known to me that many of the responsa [included in the posthumously-published eighth volume of Iggerot Moshe] were not before the eyes of my master and teacher” — or as a commenter on Hirhurim writes — “many of the later t’shuvos were ghost-written by the grandchildren based on what they thought RMF would hold.”
*This* is untrue.”
Again, your results are very problematic. In the hakdama to IM 8, in addition to the small font issue, the grandchildren admit that they wrote about 20-some of the teshuvos, and they list out which teshuvos they were. The fact that your algorithm did not detect this, again tells me that your method is not strong enough to detect such mimicry and the results are useless.
Another large issue is the way you have framed the question – “many of the later t’shuvos were ghost-written” – *many* being very vague and ill-defined, so you don’t even know what you’re trying to disprove. Again, I wrote a page-long comment about how to work around this in the blog’s previous layout, and it would be useful for you to take a look at it if it can be retrieved.
Where do you see anyone “admit that they wrote about 20-some of the teshuvos”? What they say there is that in R. Moshe’s last years he dictated some of his teshuvos (or amplifications of a note to an existing teshuvah), which were then written/typed at his direction and which he would reread for accuracy (and to which he would sometimes add notes); and then, when his eyesight deteriorated, these transcripts would be read to him and he would tell them what to correct. “Even after all this work, our teacher directed us to print only a small number of such teshuvos, those which he considered as if written in his own hand.”
(The editors’ note before OC 21, one of those “20-odd teshuvos,” adds that it comes from notes written by R. Pinchas Bodner from R. Moshe’s dictation, and which R. Moshe then reviewed/revised. In what way can that be considered something written by a different author?)
Yes, so they wrote those 20 some teshuvos and he signed off on them. The point is the same.
That will of course depend on whether they wrote down what he said word-for-word (or a near approximation thereof), or not. (The fact that R. Moshe considered some of these dictated teshuvos “as if written in his own hand” suggests that at least some of them are of the first type; people’s speaking and writing styles are often similar.)
By analogy, R. M.M. Schneerson (the 7th Lubavitcher Rebbe) didn’t transcribe his own Chassidic discourses (unlike most of his predecessors) or talks. The ones delivered on Shabbos and Yom Tov were memorized by a team of chozrim, and then transcribed as soon as possible afterwards; those delivered on weekdays were (usually) recorded on audio, and transcribed later. (In turn, some discourses and talks in both categories were later reviewed and edited by the Rebbe, and others weren’t.) But no one would say that this maamar is by R. Yoel Kahan or this talk is by R. Michoel Seligson, etc.
Irony is that the teshuvah mentioned by R Bleich was printed in Am Hatorah (Mahadura Beis Volume Ches) while Rav Moshe zt”l was alive.
Another explanation of the spike may simply be the deaths of other Great Gedolim, and thus the necessity to send all questions to Reb Moshe.
For instance: Rav Henkin ZT”L passed away in 1973, and the Satmar Rav ZT”L in 1979. Thus in 1980, for the first time, “Rav Moshe reigned alone.”
Most importantly, much of this can be clarified forever by consulting the questioners.
Time is running out; Can Nachshon step up to the plate?
[I can’t do it as I personally do not doubt any volumes of Igros Moshe-after all; Reb Dovid ZT”L and YBCL”C Reb Reuven signed on them-and for me to drive old men crazy for someone else’s doubts doesn’t sound right.]
Indeed, did no one ever simply ask the late R. Dovid or R. Moshe Dovid z”l?