Welcome! This site is a space promoting rigorous philosophical analysis of any aspect of Judaism. We look forward to your participation. THIS WEBSITE HAS MOVED! IT CAN NOW BE FOUND AT http://www.theapj.com/blog

Wednesday, 21 September 2011

The Kuzari Principle

There is an argument known as the Kuzari Principle. It tries to justify belief in whole swathes of the Biblical narrative, especially in the revelation at Mount Sinai. In this blog post, I hope to show that the argument is much stronger than it might seem. The name of the argument is slightly unfair, as it was first put forward not in R. Yehuda Halevi's Kuzari, but in Saadya Gaon's Emunot Vadeot.
Saadya Gaon starts out by noting that of all the public miracles witnessed by the Jewish people, he ‘personally … consider[s] the case of the miracle of the manna as the most amazing … because a phenomenon of an enduring nature excites greater wonderment than one of a passing character.’ He goes on to consider how the Jewish people could ever have come to believe that such a thing occurred:
Now it is not likely that the forbears of the children of Israel should have been in agreement upon this matter if they had considered it a lie… Besides, if they had told their children: ‘We lived in the wilderness for forty years eating nought except manna,’ and there had been no basis for that in fact, their children would have answered them: ‘Now you are telling us a lie. Thou, so and so, is not this thy field, and thou, so and so, is not this thy garden from which you have always derived your sustenance?’ This is, then, something that the children would not have accepted by any manner of means.
The Khazari King with whom the Rabbi debates, in the Kuzari, wants to know why the Rabbi introduced his God as ‘The God of Abraham, Isaac and Jacob,’ instead of by the more grand description, the creator of the heavens and the earth. The Rabbi states that he was following a hallowed precedent: ‘In the same strain spoke Moses to Pharaoh, when he told him: ‘The God of the Hebrews sent me to thee,’ viz. the God of Abraham, Isaac and Jacob … He did not say: ‘The God of heaven and earth,’ nor ‘my Creator and thine sent me.’ In the same way God commenced His speech to the assembled people of Israel: ‘I am the God whom you worship, who has led you out of the land of Egypt,’ but He did not say: ‘I am the Creator of the world and your Creator.’’ And the reason that the Rabbi gives for this traditional fa├žon de parle is that the Jewish relationship with God is, first and foremost, a personal one. They have known God directly through His miracles and providence, and thus, ‘I answered thee as was fitting, and is fitting for the whole of Israel who knew these things, first from personal experience [especially via the national revelation at Sinai], and afterwards through uninterrupted tradition, which is equal to the former.
What we call the Kuzari Principle is what I italicized in the quotation: Personal experience is equal to uninterrupted tradition as a ground for belief. Surely the principle is in need of some refinement, but the basic idea is true. The scientific community doesn’t feel the need to repeat every experiment ever conducted before accepting the relevant findings for themselves. If there wasn’t a modicum of trust given to the testimony of past scientists about their findings, then we could never move on, because we’d first of all have to repeat the entire history of science in our own laboratories. If we take the best elements of these medieval arguments and try to fashion them into something formal, and prima facie worth considering, we end up with something like the following:
(KP1)     If a reliable witness, called person-y, witnesses an event, event-x, and an uninterrupted chain of reliable sources passes on person-y’s testimony, even over many years, then, given certain provisos about the make-up of the chain, about person-y and about the reported circumstances of the original observation, we have good reason to believe that event-x transpired.
(KP2)     The epistemic warrant provided by such a chain increases with the number of reliable observers in the chain who claim to have witnessed the event themselves.
(KP3)     The Biblical Narrative states that the entire Jewish nation witnessed the revelation at Mount Sinai and the miracles of the exodus.
(KP4)     The Biblical and Jewish Narrative, as well as Jewish History, indicate that every generation of Jews after the exodus and the national revelation passed the story on to their children. This chain continues until this day.
(KP5)     Jewish children in every generation, given 1 and 2, have good grounds for believing that there really was a miraculous exodus and a national revelation.

The argument can also be put in the following terms:

(KP*1)      For any widespread historical belief, the belief is either true, or was sold to the general public via some sort of witting or unwitting deception.
(KP*2)      The historical belief in the national revelation at Sinai is widespread among the Jewish people, as is the belief that this knowledge has been passed down faithfully from generation to generation since the event itself.
(KP*3)      Given (KP*1) and (KP*2), the Jewish belief in the revelation at Sinai is either true, or was sold to them via some sort of deception.
(KP*4)      At no point in time could a whole nation have been deceived into the content of the story in question. A nation could certainly be deceived about ancient history, as many Britons were deceived into believing that there was a King Arthur in a court called Camelot. But part of the story that we’re talking about, as specified in (KP*2), makes the outlandish claim that every generation, including the generation that is now listening to the story, received the tradition from their parents. Thus the Jewish people would never have adopted the narrative in question, unless all of their parents had already told them the story. At no point would the Jewish nation have bought the lie that millions of their forbears witnessed something and that it was faithfully passed down from generation to generation, unless all, or most of the parents of their generation had already told them, and unless the nation was already a sizeable nation.
(KP*5)      Given (KP*4), the widespread belief among the Jewish people in the story of the revelation couldn’t have been spread by deception.
(KP*6)      Given (KP*5) and (KP*1), the widespread belief among the Jewish people in the story of the revelation must be true. The same argument will work for the story of the exodus from Egypt.
If this argument is sound, then the Jewish people have good reason to believe that there really was a miraculous escape from Egypt and a revelation at Mount Sinai.
Unfortunately, the arguments which the medieval philosophers used in order to grant Judaism its Divine sanction are open to various criticisms. I think that they can be overcome, if we’re willing to revise the argument somewhat, but first, I list the criticisms.
(P1)  Both of the arguments above only get going if we assume that the Jewish people were relating to their narratives as a history. But I often argue on this blog that the Biblical narratives are wrongly regarded as an attempt at history; that they have more in common with myth.
(P2)  We all know that traditions handed down from generation to generation are subject to corruption. The scientific community only trust their forbears to the extent that they had rigorous methods for observing, recording, and transmitting their results. This cannot be said of an ancient society. So, even if the argument can prove that something remarkable happened to an entire nation in the desert, we have no reason to trust any of the details of the story.
(P3)  Building on from (P2): It is generally taken for granted, among archaeologists and Biblical critics, for instance, that the ancient Jewish religion grew up slowly as different tribes with different religious traditions and narratives merged. As those tribes merged, the narratives merged. Thus the narratives in question are more likely to have been the product of a slow evolution than to have been the product of a reliable chain of testimony over time.
(P4)  In the book of Kings II, chapters 22-3, it is reported that a hitherto unheard of scroll was found in the temple. Many scholars think that the scroll’s appearance marks the moment that the book of Deuteronomy was first introduced to the Jewish people. The story implies that everybody accepted, uncritically, that the scroll and its content were authentic. Nobody asked how there could have been an ancient, God given, book of the Bible that nobody had told them about beforehand. This all implies that people were more susceptible to deception than (KP*4) allows for.
(P5)  The Bible and the prophets don’t try to hide the fact that at various points throughout Jewish history, idol worship was widespread, and that there was a need to encourage a return to Judaism. So a story about a mass revelation that was subsequently forgotten might have been easy to spread at some points in time precisely because idol worship had taken over for generations, and thus nobody would have been surprised that their idolatrous parents hadn’t told them.
I will respond to these criticisms in turn, and in so doing, a new and improved version of the argument will hopefully arise.
In response to (P1), I say the following. I still maintain that when faced with a narrative about his distant history, the ancient Israelite was unlikely to have been concerned, at least first and foremost, with the story’s historical accuracy rather than with its ethical force, symbolic resonance, and personal impact. This wasn’t because they had no notion of truth and falsehood. On the contrary, they must have had. Rather, they wouldn’t have worried about historical accuracy for two reasons: 1) the narratives were to be related to, culturally, as myth rather than history, and 2) the narratives in question concerned the distant past; since they didn’t have any of the tools that we now have, from archaeology to cosmology, to verify the historical claims of narrative about the distant past, they simply had to abandon the desire for historical accuracy – there was no way of knowing either way. But, some of the stories of the Bible, especially those from the exodus and onwards, wouldn’t have struck, and still don’t strike the Jewish reader purely as being about the ancient past, lying beyond the verificataroy reach of a primitive people; some of those stories don’t just posit that some events occurred a long time ago; they also posit an uninterrupted chain of transmission from then until now. I want to call that kind of story a g-narrative. G-narratives are any story with the following two characteristics:
(G1)      The events of the story were, according to the story itself, witnessed by an entire generation of an already numerous tribe/nation.
(G2)      The story itself claims that the witnesses endeavoured to initiate a chain of transmission from generation to generation – ‘And you shall tell your child’ – such that their progeny would never forget what they saw.
Jewish narratives are treated, by the religion, as myth – given that it bases rituals, retellings and re-livings upon them. But, myths can be based upon historical facts. The truth or falsehood, in terms of historical accuracy, of a myth is not a key factor in the worth of that myth. But, when some of the historical claims of a myth do lie within the verificatory grasp of a culture, that myth is unlikely to be adopted, unless its historical claims were seen to have stood up.  Once a myth is adopted, however, and embedded into a religious culture, like the myth of the Garden of Eden, for example, its historical accuracy is no longer relevant to its worth as a religious myth.
G-narratives, even when their storylines are set a long time ago, are not completely beyond the verificatory reach of their audience, because, at the very least, the audience will know whether the story was passed down to them by the entirety of the previous generation. G-narratives in the body of Jewish narratives are an exception, rather than the norm. But they are significant for the following reason: it’s difficult to believe that any g-narrative could have become widespread as a myth unless it had been passed down from generation to generation. The only generation that would have accepted a g-narrative, even as the basis of a myth, or so it seems to me, would be the generation that witnessed the key events of the story for themselves. Otherwise, they’d have said, ‘I didn’t see this, and no one ever told me about it before now!’ If a g-narrative is widespread across a culture, however primitive, you’d have good reason to believe that its central story line, or something similar to it, actually occurred. To accept this thesis is not to revert to anything like an historiographical approach to Jewish narrative in general.
It’s also worth pointing out just how few of the central Jewish narratives are g-narratives. All of the stories that pertain to times before the emergence of the Jewish people fall short of being g-narratives. All of the narratives that weren’t supposed to have happened before the entire nation also fall short. There are even narratives that talk of miracles that were beheld by the entire nation, such as the sun standing still for Joshua, that fail to be g-narratives because the story doesn’t talk of an endeavour to pass the story on from generation to generation. The story of the exodus from Egypt and the mass revelation at Mount Sinai, which are both accompanied with commandments to keep the story alive, are some of the rare occasions in which a Jewish narrative can claim to be a g-narrative. In fact, very few religious or historical narratives outside of Judaism can claim to be g-narratives either.
In response to (P2), I have to make some concessions. It’s true. A narrative transmitted down the generations can be subject to all manner of subtle and gradual evolutions. Imagine a massive intergenerational game of Chinese Whispers. By the time that the message has reached the end of the line, its originator has died, and can’t tell you whether its integrity has been preserved. Nevertheless, one would imagine that the main headlines of the story, even if all of the details became perverted, would remain constant down the chain. Any major rupture or change to the general plot-line, and any diminution of its most striking claims, would surely be noted by a public in love with its legends and folklore. But, we can imagine a gradual process of exaggeration, and we can imagine a gradual corruption of mundane details. For that reason, a widespread g-narrative cannot be trusted as an accurate history. Its religious function, thankfully, isn’t to serve as an accurate history – because as I say, historical accuracy isn't the purpose of myth. But, one can trust, or so it seems, that its main claims must have had some basis in fact, however tenuous. An uninterrupted chain of testimony doesn’t have the truth-preserving qualities attributed to it by Saadya Gaon and R. Yehuda Halevi, but it does justify the less ambitious claim that the story must have had some sort of grounding in fact. The story of the exodus and the revelation at Sinai is now a widespread g-narrative. This seems to justify the claim that an entire nation or tribe had some remarkable religious experiences, once upon a time, even if the finer details of the story can’t be trusted as an accurate historical account.
I don’t want to get into the whole question as to the reliability of Biblical criticism and the historical claims that are made in its wake. Inspired by Umberto Cassuto, I have my doubts about Biblical Criticism as a science. Notwithstanding, I think that the whole question is largely irrelevant to the philosophy of Judaism. Traditional Judaism relates to some of its texts as if they were written, word for word, by God Himself. The philosophically interesting question is: what does that attitude towards that text achieve; what effect does it have; why and how? Even though I’m dubious about the findings of Biblical Criticism, and dubious of their philosophical relevance even if they’re true, I would like to assume, only for the sake of argument, that the Biblical Critics are right about the way in which the Jewish people slowly grew into one nation via an amalgamation of different tribes; that each tribe brought with it its own variously divergent religious narratives and traditions. I contend that even if we accept this assumption, a version of the Kuzari Principle can still get through. This would show, in turn, that (P3) simply isn’t worth worrying about.
A tribe is unlikely to accept a g-narrative if they didn’t witness the events of the narrative for themselves, or if they didn’t inherent the narrative as a tradition from the previous generation. But, there is one important class of exceptions. At any point in the life of a tribe, outsiders can join that tribe. When they do, they often adopt the history of their new found tribe as their own. As a British Jew, I was taught British history, even by my parents, as if that history was my own, even when the historical events in question happened long before my ancestors became British. I have had conversations with American Jews in which they mock me and my kind for having lost the war of Independence against the Americans in 1777, but, in 1777 both their ancestors and mine were living in Eastern Europe! When outsiders join a tribe, they adopt its narrative as their own.
Let us imagine that tribe A has a g-narrative. Let us imagine further that members of tribe B or even the whole tribe joins tribe A. At that point, members of tribe B, who have joined tribe A, start to teach their children the g-narrative of tribe A as if it were their own history. It is still true to say that tribe A is unlikely to have had a g-narrative in the first place unless they had witnessed its key events for themselves, or if they had inherited the narrative from the generation before them. Thus, it’s still true to say that the existence of the g-narrative is good grounds for believing that its content is in some way grounded in fact. Even if new members of the generational chain of transmission (henceforth, the g-chain) find their way into that chain somewhat artificially. Nothing in this example is changed when two tribes merge and decide to adopt one another’s narratives. When the narratives are g-narratives, their very existence, which allows them to play a role in the inter-tribal merger, is good evidence that their content is somehow grounded in fact.
Thus, even if the Jewish people and their narratives emerged after a long process of intertribal merging, the presence of g-narratives in the resulting tradition is still good evidence that those narratives are grounded in fact. For that reason, it seems that (P3) is largely irrelevant to the Kuzari Principle.
In response to (P4): It is true that 'a book of the Torah' was discovered in the temple during the reign of King Josiah, according to Kings 2, chapters 22-3. Let us concede for the moment, that the book discovered was the book of Deuteronomy, or some forerunner to that book, as is the consensus among Bible scholars. It is interesting to note that the book of Deuteronomy, though it contains mentions of certain g-narratives, isn't the source of any of them. The book of Deuteronomy is generally just a re-cap of the previous books of the Bible, of both their narrative and legal sections.
There are, admittedly, a few new laws in the book, but from the context of the story of its 'rediscovery' in the temple, we see that King Josiah was impressed not so much with new information, in terms of new stories or new laws. What impressed King Josiah about the book seems to have been, if it was the book of Deuteronomy, the stirring words of rebuke that Moses delivers in it to the Jewish people. Moses warns the Jewish people, in the book of Deuteronomy, that if they sin, and follow other Gods, as they were doing in the time of King Josiah, that there would be great suffering and destruction. It was this information, more than anything else in the book that seems to have moved King Josiah to tears, and to renting his clothes. When King Josiah declares: '[G]reat is the wrath of the LORD that is kindled against us, because our fathers have not hearkened unto the words of this book, to do according unto all that which is written concerning us,' he sounds like a man who's had the fear of God driven in to him by the fearful threats of Moses in the book of Deuteronomy.
The 'discovery' of this book of the Torah is irrelevant to the Kuzari principle for a couple of reasons. Firstly, if it was the book of Deuteronomy, then it doesn't bear any relevance to Jewish g-narratives, as none of them originate in the book of Deuteronomy. Secondly, it seems that what the discovery of the book inculcated, consistent with it's being the book of Deuteronomy, was a new found religious fervour rather than a new found religion!
In response to (P5) I make the following point: the Biblical account of Jewish history, consistent with everything we know about religiosity in the ancient near east, paints a picture of a people in which syncretism and monolatrism were generally rife. If this is the case, then just because belief in and worship of other gods may have waxed and waned, there is no reason to think that knowledge of, and belief in the God of the Hebrew Bible, and the associated g-narratives ever disappeared. Instead, the God of the Hebrew Bible was often regarded, by the Jews of the Bible, as one of many regional gods competing for their affections. Syncretism and Monolatrism were the bane of the prophets' existence. 'How long will you waver between two opinions? If Hashem is God, follow him; but if Baal is God, follow him,' declared a frustrated Elijah who wanted the Jewish people to make their mind up. But there is no notion that Hashem had been forgotten about at any point, or that the transmission of g-narratives down the g-chain had ever been threatened. On the contrary, the Biblical critics would have us believe that the various tribal traditions and legends that went into the formation of the Bible stretched back into the very distant past.
The Kuzari Principle had to face off five major concerns. Having looked at each of these concerns in turn, it looks like a version of the argument can still survive:
(rKP1)       A myth will not initially find any traction with a culture who know it to be historically inaccurate, although once it has been adopted and becomes culturally significant, then a culture will not abandon it merely because of its historical inaccuracies.
(rKP2)       The generations of a g-chain will not accept a g-narrative about them unless:
a.       The generation is the first in the g-chain and witnessed the events for themselves, or
b.      The generation received the g-narrative as an inheritance from their parents, although each generation after the first may have members outside of the core of the g-chain who adopt the g-narrative as they join or merge with the community and adopt its history as their own, or
c.       The entire generation becomes convinced, presumably only with a large amount of evidence, that a g-chain that was supposed to have reached them was broken – this doesn’t seem to be the case with the small number of Jewish g-narratives, all of which seem to stretch right back into the mist of Jewish pre-history.
(rKP3)       G-narratives can be subject to slow corruption and exaggeration. Thus, the fact that a g-narrative is widespread among a contemporary generation of a g-chain is no reason to believe that that g-narrative presents an accurate history.
(rKP4)       The original g-narrative, before any corruption or exaggeration, must have been sufficiently impressive in and of itself to have initiated a sustained desire to transmit the story down a g-chain.
(rKP5)       There are a small number of remarkable and widespread g-narratives about the Jewish people, about mass revelation and divine deliverance. These narratives are still transmitted to the majority of Jewish children by their parents at events like the Seder Night, in which Jewish parents recount the exodus from Egypt for the benefit of their children. This ritual retelling is even conducted by a very large number of Jews who no longer believe in the historical accuracy of the narrative.
(rKP6)       Given (rKP3), the fact that these g-narratives are widespread among the living members of the Jewish g-chain is no reason to trust the historical accuracy of the stories.
(rKP7)       But, given (rKP2) and (rKP4), the fact that these g-narratives are widespread does provide good reason to believe that the story wasn’t initially adopted by a generation to whom the story didn’t actually happen, and that the story is grounded in fact even if it’s been distorted, and that the facts, in and of themselves, were sufficiently impressive to generate the very long lasting feeling of cultural obligation to pass the story on down the g-chain.
(rKP8)       We have no reason to believe that the details of Jewish g-narratives are historically accurate, but we do have reason to believe that they are grounded in extraordinary facts witnessed by an entire generation of the g-chain.
(rKP9)       Therefore, we have good grounds to believe that an entire generation of the forbears of the Jewish people, or an entire generation of a tribe that would later amalgamate into an emergent Jewish people, were collectively witness to an extraordinary sequence of supernatural events, and events that would have been collectively understood as Divine revelation.
We dismiss (P1) with (rKP1). (P2) is given due weight by (rKP3) and (rKP4). (P3) is duly noted by (rKP2b). (P4) and (P5) really shouldn't bother us, for reasons I went into above, but the revised argument alludes to them with (rKP2c)
I contend that this revised Kuzari principle makes it rational to believe that the forbears of the Jewish people experienced a mass revelation of Divinity and that there was a miraculous Exodus from Egpyt, even if the numbers involved were very much smaller than in the Biblical account. I'm interested, as always, to hear people's responses!


  1. Sam, this is interesting and the analysis of the argument is helpful. I'm still not convinced of its soundness though.

    Couple of points:

    The King Arthur counterexample is a good one, and I don't think KP*4 does the work necessary to defeat it.

    KP*4 assumes that points G1 and G2 are true, but also that the tribe came to believe G1 and G2 at the same time.

    What if a tribe somehow started believing a G1-type claim like Arthur, which as we've seen can happen (some Greek myths might also qualify)? Every generation receives this tradition from their fore-bearers. At some point, people come to believe that the act of passing on the story is a part of the story itself. Maybe it's not explicitly stated, but the way the story is told makes it seem that the telling itself is significant. At this point, the tribe would start believing G2 but wouldn't run into the sorts of problems that are entailed by G2 being in the picture at the time that the story first came to be believed.

    This seems like a quibble, and it is. I suspect that with a bit of time I could find a couple more, though you might be able to fix them. But actually I'm a little distrustful of the project as a whole here, applying analytical tools to the argument. The Kuzari principle's flaw is it tries to apply logic to an array of highly complex sociological phenomena. We don't really understand things like cultural memory and the development of mythos. We don't really understand how ancient Israelites saw the world. I'm not sure analytical tools can do very much work here at all.

  2. Thanks Arieh, you point about the time at which G1 and G2 are accepted is a facinating one. I have fleeting feelings that its deeply relevant and potentially destructive to the argument, but then fleeting feelings that it can be overcome, or that its not as bad as it sounds.

    Either way, it's a good point and I need to think about it.

    I don't think there's anything inherently wrong with applying logic or analysis to this general topic, but your last paragraph still gets right to the heart of the matter. (rKP1)simply assumes too much about the processes by which myth was adopted in ancient cultures. This doesn't mean that the argument isn't sound, but that it hasn't yet proven to be sound. That is to say, we need more data. We need to know from ancient historians and anthropologists and the like, a lot more about these topics. But if we found out that (rKP1) and (rKP2) were true, then the argument would have a lot of force, or so it seems to me.

    Finally, its worth noting, that the argument never sought to prove beyond any doubt that certain historical events occured; it rather seeks to provide epistemic warrant for a certain set of beliefs about the past. So, the argument isn't as ambitious as some people make out.

    As always, thanks for your thought provoking arguments (and if I ever publish anything on this, I'll attribute the King Arthur example to you!!).

    Shabbat Shalom

  3. I've had a little think about your first point. Now I know why I both agreed and disagreed with it!!

    It seems to me that, on the one hand, you're right. Conceivabley, G2 might only become a proper part of the story many years later, which would cover up a potential gap in the transmission of the story.

    But, for any given G-narrative, that would be an empirical question - what sort of gap was there between the adoption of G1 and G2? And, how can we know, even if there *was* a gap, that the transmission of the core narrative nevertheless stretches so far back in time that it could only have been adopted by a people who knew that it was true.

    Even according to the Biblical Critics, the story of the exodus (at least the G1 part) was known to the tribe/tribes that the story was about, already in the pre-history of the Jewish people.

    This again increases the likelihood (because increasing liklihood is all that kuzari principle is good for) that the story streches back as a cultural narrative, pretty much all the way back until it was alleged to have happened; making its initial adoption unlikely, unless the story was true (albeit the story before it was exagerrated and corrupted over the generations), even if G2 was, itslef, a later addition to the story.

    I'm not being very eloquent here. Do you get my point?

    So, once again, we come back to you last point: we need to know a lot more about ancient Israelite culture and the birth of mythology in general. Notwithstanding all that you say, the Kuzari argument certainly could prove its conclusion - that we have epistemic warrant to believe in some sort of Divine interaction between the forbears of the Jews and God. All we need to know is how far back in time people were telling G1. G2 is simply a way of making sure that the narrative stretches all the way back. But if G2 is found to be a later addition to the narrative, the argument can still work, as long as we have evidence that G1, by itself, streches far back enough in time, as a traditional narrative, to ensure that its origional tellers had to have known whether or not the story was true.

    Does this make sense?

  4. FREE INTERNET DOWNLOAD; "Man's Search For Spirituality by E Christopher Reyes.
    A chronological history of religious beliefs.
    E Christopher Reyes

  5. I'm not sure the first part of (rKP1) is acceptable. Can it be rephrased, as in "A myth will initially find traction within a culture that thinks it to be historically accurate"?

    Now the issues are clearer. What is the basis for determining why myths do or do not gain traction in a culture? The answer--whatever it is--comes from historical studies and interpretations of data. The answer is not one that comes very well from philosophy, or at least philosophy alone.

    My whole series of Kuzari posts comes back to this basic fact, that no "principle" itself carries much weight in helping to answer historical questions. As I have asked with respect to the alleged revelation at Sinai, "if we have no evidence today of the single most extraordinary thing that's ever happened in the history of the universe, how much credence is appropriate for me to give?"

  6. Larry,
    Thank you so much for the comment.

    Yes, I can accept the amendment to (rKP1) though I'm not convinced that it's totally necessary.

    The basic fact that you say you keep coming back to is certainly not in dispute by me. I agree: No principle itself carries much weight in helping to answer historical questions.

    In many areas of philosophy, the philosopher has to wait to see what other disciplines discover. Think for instance of the effect that Einstein's theories about space and time have had over metaphysics. Philosophy has to be in conversation with other disciplines.

    I am convinced that (rKP1) could only be made plausible by a great deal of historical and anthropological data and educated interpretation of that data.

    All of that being said, I'm merely interested to find out the following from you:

    1) Do you accept that my reformulation of the Kuzari argument (let's call it that, instead of the Kuzari principle), is better than other formulations? Better in terms of having fewer obvious holes and flaws, and in terms of being more persuasive, even if it doesn't actually convince you.

    2) Do you accept that if all of the premises of the argument, as I have reformulated them, were shown to true (with the help of historians and anthropologists), then you'd actually have good reason to believe that something like the revelation at Sinai happened even though we don't have any 'direct' evidence of it?

    As far as I'm concerned, that is the only power that the Kuzari argument could possibly hope for: the power to justify our belief in Sinai, or something like Sinai, without direct historical evidence. But you'd still need something like 'indirect' historical evidence; i.e., evidence about how myths were formed in ancient times. I'm not claiming that we can arrive at justified historical beliefs a priori.

    I look forward to continued discussion with you, and to your answers to these questions. I'm certainly far from convinced that the Kuzari argument is even as strong as I suggested in this post. So, the discussion is really important to me. Best wishes.

    1. This comment has been removed by the author.

    2. Hi Sam,

      "1) Do you accept that my reformulation of the Kuzari argument (let's call it that, instead of the Kuzari principle), is better than other formulations?"

      No, I do not. You are not working in logic but in human behavior, and human social behavior at that.

      "2) Do you accept that if all of the premises of the argument, as I have reformulated them, were shown to true (with the help of historians and anthropologists), then you'd actually have good reason to believe that something like the revelation at Sinai happened even though we don't have any 'direct' evidence of it?"

      I'm sorry, but no. What's wrong with Kuzari is its approach more than anything else. That approach, by its very nature, is incapable of providing sufficient justification for a Sinai event.

    3. Thanks Larry.
      Let me respond to your two points.
      1) I don't think I understand how you're trying to utilise your distinction between logic and human behaviour. According to most logicians, one of the key distinguishing factors of logic is its topic neutrality. That means that every area of discourse should be subject to the domninion of logic - even discourse about human behaviour - even if human behaviour sometimes exhibits irrationality, talk about human behaviour should still be governed by logic. Perhaps I could understand your point better if you could point to the logical fallacy that I make, or the false premis about human behaviour that I rely upon.
      2) I don't want to claim that the Kuzari argument proves or makes more likely that the revelation at Sinai happened, but that SOMETHING like the revealation at Sinai happened. As Yoni points out below, this could merely mean that there was a remarkable natural occurance witnessed by a generation of the forbears of the Jewish people.

    4. Sam,

      I’ll try to explain why the distinction between logic and human behavior is important. You accepted my reformulation of rKP1, which I gave as "A myth will initially find traction within a culture that thinks it to be historically accurate." My point is that this statement cannot be true without sufficient historical evidence to support it. At the very least, I think there is probably a way to get some small-scale experimental validation of the premise. To my knowledge, however, there are no historical or experimental supports for the premise.

      Moreover, the premise is not logically sound. Consider this alternative premise: “A myth will initially find traction in a culture even though people know it isn’t historically accurate.” Or consider this one: “A myth will not initially find traction in a culture that thinks it to be historically accurate.” Of our three premises, is only one correct and the others false? How do you know? Are they all true some of the time? How do you know?

      My point is that I see no justification to accept rKP1, and I see no reason to base an argument on it. You may want to use Kuzari to suggest a sufficient likelihood that SOMETHING Sinai-like happened, but I don’t think Kuzari even allows that.

    5. Thanks Larry.

      I think one of the reasons I'm having touble understanding you, and this sounds condesending but I'll explain what I mean by it in a minute, comes from the fact that I'm a classically trained logician.
      This means that I use logical terminology in a very strict and restricted way.

      Just because a premis of an argument depends upon empircal evidence doesn't mean that the argument itself isn't logical. Take the classical example of a logically valid argument:

      1. All men are mortal
      2. Socrates is a man
      3. Therefore Socrates is mortal.

      This argument can never be known to be true until we have good reason to believe the first premise. The first premise isn't logical in its content, it's biological. The argument also relies upon the truth of the second premis. The content of that premis isn't logical either, again, it's biological. But, logic dictates that, given the truth of the first two premises, the conclusion will follow. That’s all that interests the logician. That’s why we call it a logical argument despite its biological content.

      The study of logic allows us to track the validity of the argument but not its soundness. For all we know, this valid argument may be false. Perhaps Socrates was actually an imortal robot planted on the earth by aliens. Or, perhaps medical science will one day reveal that not all men are mortal. Such a discovery will just stip the premises of their apparant truth, but it won't strip the argument of its logical validity.

      Likewise, what you're pointing out is that rKP1 is ABOUT human behavour. It's content dosen't concern logic; much like the premises of the Socrates argument. I readily accept that rKP1 would need to be substantiated by anthropologists and by other experts with a sufficient data base to establish the truth or falsehood of the claim.

      So, let me rephrase my question to you.

      If you can be convinced that rKP1 is true, by the relevant authorities appealing to the relevant data, do you think the argument follows to the conclusion that SOMETHING like a Sinai experience must have happened? Just like, if you can be convinced that Socrates really was a man, and that all men are mortal, you could be convinced of the conclusion that Socrates was mortal.

      If not, if you still wouldn't accept the conclusion, I'd like to know why, because that might indicate your feeling that there's an actual flaw in the logic of the argument. That would interest me.

      Thank you for discussing this with this me.

    6. This comment has been removed by the author.

    7. Sam,
      I appreciate your discussion of how you are approaching the logic--by which I mean, generally, both the validity and soundness--of your Kuzari premises.

      My point to you is that rKP1 cannot be shown to be logically valid or logically sound. It fails on both. I imagine you appreciate my position on why it fails on soundness: it’s probably not true. To me, this fact overwhelms the question of validity; after all, if the premise is false, who cares if it’s valid?

      But you seem to want to talk about validity, so let me emphasize: the logical validity of rKP1 certainly has not been demonstrated to my satisfaction. You may have explained yourself on this point, and if I have missed it I apologize. If not, can you please show me on what basis rKP1 can be considered logically valid?

      When I look at what you have in your original KP1, and then onward to rKP1, what I see is--forgive my bluntness--sloppy thinking. Most every key term in that premise is vague and/or untethered from the reality of what happens in actual life. For example, in KP1, what exactly makes one a “reliable witness”? How is that determined? Are we talking about someone with a complete and accurate view of how something actually happened, or are we talking about someone who was generally known to be honest in all situations?

      To me, KP1 and rKP1 suffer from an infestation of baggy terms that, taken together, make the premises almost impossible to reconcile with reality.

      Please note the contrast between the Kuzari first premise with your “All men are mortal” example. Categories such as “men” and “mortal” are clear enough for me to accept them. I am comfortable with these categories in a way that I am not with terms such as “reliable,” “witness,” “myth,” “passes on,” “traction,” “culture,” and so on. So, while I can accept “All men are mortal” because I have clear-enough terminology and sufficient observational/experiential data, neither of these conditions are met in rKP1. As far as I’m concerned, rKP1 cannot be considered valid until its language is refined enough to allow me to agree that it is true under every interpretation. In other words, I can agree that “All men are mortal” is true under every interpretation. I cannot agree the same for rKP1.

      Your task, then, is to work rKP1 until it is indisputably valid. I think Rabbi Dovid Gottlieb does a nice job with this; unfortunately, it doesn’t help his case in support of Kuzari.

      Finally, you ask “If you can be convinced that rKP1 is true, by the relevant authorities appealing to the relevant data, do you think the argument follows to the conclusion that SOMETHING like a Sinai experience must have happened?”

      I imagine that if you can get rKP1 to a point of validity, then that will go a long way to helping clarify subsequent premised. You may not know that I have offered an opinion on SOMETHING like a Sinai event happening. Here is the link (http://larrytanner.blogspot.com/2011/07/how-sinai-story-originated-and.html), and here is what I said:

      “So we can only speculate…that the Sinai story was a fabrication using pre-existing narrative elements which may have been factual at some point in the distant past. The story was part of a larger tapestry of narratives involving legendary figures and the local god. The story later became taken as historical. It became the basis for justifying the various daily religious observances and obligations of the people.”

      In the end, I think you and I get to a point of some agreement.

    8. Thanks Larry,

      Just a couple of points to make in response.

      1) You're still not using the word 'valid' in the traditional sense, which leaves me with some trouble understanding what it is you're getting at. A single premis of an argument can't properly be called valid or invalid. Validity is a relationship between premises and a conclusion. So rKP1 just can't be spoken of in that way.

      And I simply disagree that validity is uninteresting when the premises are false. Validity generally interests the logician much more than soundness; the mere question as to whether the premises are true! But that's just a disagreement of taste.

      2) What I can make out of your grevance with the argument is the vagueness of the terminology it uses. I can make a couple of quick responses to this discomfort you're having:

      a) vagueness of the terminology cannot affect the validity of an argument as long as the terminology is constantly used with the same significance, and the contours of its vagueness stays the same on each of its uses in the argument.

      b) What number of cells does it take to make a man? Mortality is ambiguous because our definitions of death differ. Outside of mathematics, you'll have trouble finding any argument that isn't infected by a good dose of vagueness and/or ambiguity. But we manage to get by!

      I'm not sure this argument suffers as long as we all have similar understandings of the key words, just as we have similar understandings of the word 'red' or 'heap', 'man' and 'mortal', dispite lacking formal criteria for the appropriate use of these words.

      That being said, I'm all in favour of working together on definitions of 'witness', 'reliablity' and the like, which is the job of epistemology; and 'myth' and 'cultural traction' and the like which is the job of social philosophers.

      I agree that our conclusions aren't all that far apart in the end.

    9. My problem with Rabbi Gotleib's take on the argument is that he thinks that the premises necessitate a certain conclusion about sinai itself.
      I think that that's too strong.
      I think they make a certain conclusion about Sinai more plausible. Thus, if my argument is valid, the premises, if they were true, would merely necessitate the following sort of conclusion: We have good reason to beleive that X.
      Rabbi Gottleib is much more ambitious, and I think its a weakness.

    10. Sam,

      The heart of our disagreement is this, where you say: "I'm not sure this argument suffers as long as we all have similar understandings of the key words."

      My criticism boils down to the fact that no one is establishing the understandings of the key words. The understandings of these words are precisely the issue.

      Let me back up, however, my more important criticism is that where we have facts and data about matters similar to those presented in rKP1--the facts and data falsify rKP1.

      But if we decide not to care about facts, my question is how you can ask me to accept rKP1 when I do not know what it means. You say that the incomprehensibility of the premise is no problem for the larger argument. You also say that lack of data supporting the premise is no problem.

      So I guess I am really at a loss to figure out or remember what it is we're doing. May I ask: do you find any premises problematic, ever? If so, what triggers it for you?

      If we're going to be completely academic, I suppose I can accept rKP1 in the same way I can accept the premise "All men are 17-meters tall." If our terms mean everything and facts mean nothing, how can I do otherwise than say "lay on, MacDuff"?

      Please humor me and see if we cannot get a bit more refined in the terms of rKP1. What happens if you try to translate some of the premises symbolically?

    11. Larry,
      I certainly didn't say that incomprehensibility of a premise is inconsequential to me.

      I also didn't say that lack of supporting data is no problem?

      Where did you get that from?

      My point was that, even if there are vague terms, then an argument can still be fine, as long as we all know the vague meaning of the vague terms.

      I said that lack of supporting data is no obstacle to the validity of an argument, but it is an obstacle to its soundness, and of course, ideally, I'd want the argument to be sound, even though validity here would still be an impressive finding.

      Regarding a symbolic translation of rKP1, any sentence of a natural language can be translated into some sort of symbolic language. What would be the point of doing that here and now?

      Finally, if you do have data that disconfirms rKP1, then the argument would be forever laid to rest. So, I'm interested. What is the data?

      Once again, thank you for engaging with me on this.

    12. Sam,

      On incomprehensibility and lack of supporting data: The reason these don't seem to be a problem for you is that you continue, in my opinion, to wave away my fundamental point:that rKP1 is too vague to be useful.

      My whole series on Kuzari focused on trying to make the agument (Gottlieb's) concrete. One historical example I used was the bombings of Hiroshima and Nagasaki: http://larrytanner.blogspot.com/2010/09/going-nuclear-on-kuzari-principle.html.

      Forgive the long quote from that essay that I'll give now, but the example of the bombings showed "that Kuzari is too vague and speculative to tell us much about Sinai or belief in Sinai. In the case of Hiroshima/Nagasaki, the formation of beliefs started with confused reports of an event witnessed by very many. Information was then gathered and consensus opinions began to be formed. Later, the event was put into historical context and its significance contested.

      With Sinai, we have one report given from one perspective. We don't know the report of the people closest to the mountain. We don't know the observations of witnesses in the very back. We cannot hear the voices of the women, the outsiders, and the opponents. We have no documents from the nations closest to Sinai telling us about something most unusual having happened. The strongest inference we can draw from Sinai is that something--a natural event or some other fantastic occurrence--may have happened out in the wilderness. We may even be able to justify saying something must have happened. We cannot, however, say with any confidence that Sinai happened. Kuzari changes nothing about this because even if the principle itself is 100 percent true, it doesn't tell us anything about what exactly happened at Sinai, how, to whom, over what time period, and at what stakes. It's a believer's reason to believe, a 'nice to have.' But it's not especially compelling to a neutral observer."

      The reason I'm pushing you to a symbolic translation and to get more concrete with the language of rKP1 is that I can think of no other way to help you see why Kuzari is so problematic. When you go from abstract and vague terms to a hypothetical or hstorical example, then the problem of vague and abstract terms becomes acute. It's this problem that, in its resolutions, puts the soundness of Kuzari-based arguments in grave doubt. Gottlieb had a valid argument, but it ultimately failed (or, in another view, its success was too limited to be consequential).

      In my Kuzari series, I presented data that I think would challenge, if not disconfirm, rKP1. Off the cuff, I think England's King Arthur, probably a complete fiction, might work toward disconfirming rKP1. It was not historically accurate, ever, yet the culture accepted it. And what do you make of gossip and urban legends? What do you make of 9/11 truthers and such?

      Here's what I'll do, since I am not great at communicatingin comments: I'll go through your entire syllogism and give you my notes and thoughts.

      Finally, let's be clear that I have been willing to grant that Sinai, for example, likely has its origins in some historical event: a slave revolt, a fantastic storm, the coincidence of fervent prayer and strong weather. Who knows what? But I think modern biblical criticism and modern understandings of historical and physical reality make it clear enough that there never was any sort of supernatural (i.e., divine) communion or event at Sinai.

      Thanks for the dialogue. I hope my points push you do some more work on Kuzari. As I said, I will do my best to analyze your rKP argument as carefully as I can (I'll post to my site).


      "Larry Tanner"

    13. As promised, my analysis of rKP, offered with sincere respect as well-meaning criticism.


    14. Thanks Larry for the analysis. In essence I agree with you.

      While I'm not sure that biblical criticism and ancient history need be all that relevant to my religious life, as my religious life might be about engaging with the first five books of the bible, as if they were written by God. I can still accept that if one is interested in proving the Kuzari principle to be sound, one will have to prove that all of its premises (including rKP1) are true. That cannot be done without offering rigiourous definitions of the key terms and engaging with source criticism, archeology and other relevant sciences to prove or disprove the premise.

      But, the first thing we have to do is to find out whether the argument is valid (which is to say, whether the truth of its premises would be enough to entail the conclusion). If it's not even a valid argument, then there's not much point going to the trouble of proving its premises true or false.

      That much, making the argument valid, I think I'm able to do, although the conclusion is less exciting than Rabbi Gottleib's. I think you might even agree with me that I've made it valid. But you go further and argue that its first premise is vague and most likely false even when sharpened up.

      I think that question would need much more research in diciplines that are beyond my expertise.

      On another point: I don't think that translating the first premise into symbolic language would help you at all. The whole idea of moving from natural language to symbolism, at least in logic, is when you don't really care about the meaning of the premises, but are interested in sketching their structure. It's a tool for tracking structural valditity without looking at content or truth.

      I take the argument:
      1) All men are mortal
      2) Socrates is a man
      3) Socrates is mortal

      I translate it into:
      1)(x)(G(x) > F(x))
      2) G(s)
      3) F(s)

      The symbolism moves us away from manhood and mortality and looks to the structure of the argument.

      What you're really looking for, is a move in the opposite direction. You don't want to move away from terms like 'myth' and 'cultural traction' - you want more rigorous definitions.

      I agree that we need them. But symbolism won't help here.

      I also think that we don't even need what you're looking for in order to discover that the argument is valid, which the first step to discovering whether it is sound. Now I need to pass the project on to those who have the requisite skills and data set to better define and evaluate the truth/falsehood of the premises.

    15. And I think we've come to some agreement, although I'm certainly and obviously not up for the kind of philosophical analysis that would be really useful to you.

      One question that remains for me is the where the "extraordinary sequence of supernatural events"--particularly the "supernatural" part--comes into play beforehand. I suppose there's some basis for something supernatural, specifically, in rKP4 and rKP5, but the basis doesn't strike me as forceful.

      I hope this is a constructive comment!

    16. I didn't mean to undermine your philosophical acumine at any point. In fact, your online work on the Kuzari is actually some of the best literature available on the topic.
      I take your point that 'supernatural' is smuggled in. In order to legitimise that, I'd have to do some more work, appealing to premises about how unlikely a nautural occurence is to trigger a completely new religion. But, once again, a premise like that would need a lot of data to establish as true.
      Some of the pressure is taken off by the fact that my conclusion only tries to determine liklihood or plausibility rather than historical fact. But I take the point.

  7. Thank heavens I found this!
    I've been looking for a more rigourous analysis of the Kuzari argument for months now!
    Also going to take this material for a seminar I'm running for hineni.

  8. Can you check out my blog and tell me what you think of the video in my first topic?

  9. Yechonan, can you post the url?

  10. Sam, thank you for your post. I learned a lot from it.

    You write: "But, one can trust, or so it seems, that its main claims must have had some basis in fact, however tenuous. An uninterrupted chain of testimony doesn’t have the truth-preserving qualities attributed to it by Saadya Gaon and R. Yehuda Halevi, but it does justify the less ambitious claim that the story must have had some sort of grounding in fact."

    Why is this not compatible with the claim that a rare natural event occurred at Sinai (a massive earthquake with thunder and lightning) which was construed by a charismatic leader (Moshe) as being God's attempt to communicate with the nation? Remember that this nation consisted of people who in all likelihood didn't bifurcate the world between the natural and the supernatural (Egyptologist Bruce Trigger: "the natural world was seen as suffused by supernatural energy which endowed trees, animals, rocks, and stars with reason, emotions, power, and will. This made it possible for human beings to interact socially with the natural world in the same manner as they did with powerful, and hence potentially dangerous, human beings") and therefore wouldn't greet such a claim with the same skepticism that we would.

    Again, even if I accept your (rKP8) - "we have no reason to believe that the details of Jewish g-narratives are historically accurate, but we do have reason to believe that they are grounded in extraordinary facts witnessed by an entire generation of the g-chain" - as far as I can see, (rKP8) is completely compatible with the story I just suggested and doesn't entail any type of revelation at Sinai and therefore doesn't entail (rKP9). Why should we think that these "extraordinary facts" involved revelation?

    1. Thanks Yoni.
      I think you're right, in a sense.
      What the Kuzari argument seems to justify, or make more likely, is the claim that something extraordinary was witnesses by the forbears of the Jewish people.
      This falls a long way short of saying that it was God giving any sort of Divine stamp of approval the Jewish religion that would evolve from that encounter.
      Instead, it could have been a once in a life time natural occurnace. Or, it could have been a mass hallucination caused by narcotics - as argued by Prof. Benny Shannon http://www.haaretz.com/print-edition/news/researcher-moses-was-tripping-at-mount-sinai-1.240582
      But we have no evidence that ancient Jewish rites ever involved drug taking in the way that occured in South-American tribes. So I don't find that likely.
      I'm also unaware that a religion was ever founded in response to a natural occurence, however awesome and misinterpreted it might have been.
      As a person who has had weak varieties of religious experience, it becomes easier to claim that, in all liklihood, the experience in question was a religious one; an experience of Divinity.
      But the Kuzari argument, or so it seems to me, should only be used to make the relatively weak claim that 'something extraordinary happened at some time!' - which is compatible with all sorts of disambiguations.
      Even so, the Kuzari argument is still important, because it's interesting to see how the mere existence of a widespread belief really can have some epistemic significance.


Please include your name at the end of your post or comment

Note: only a member of this blog may post a comment.