Balak (5764)

From the book Mida Tova: Articles on the Hermeneutical Principles by Rabbi Michael Avraham. Translated from Hebrew using gpt-5.4 (reasoning_effort=high, batch API).

---

Midah Tovah — Eve of the holy Sabbath, Parashat Balak, 5765

Questions

1. What is notarikon, and from where is it derived, if at all?
2. Is it remez (allusive hint) or drash (interpretive derivation)?
3. What is the relation between the letter and the word hidden within it?
4. How is all this connected to the laws of writing on the Sabbath?
5. What is the relation between the hermeneutical principles of aggadah (non-legal rabbinic teaching) and those of halakhic interpretation?
6. What is holy about the Holy Tongue?
7. How is all this connected to Kabbalah?

The Principles

• Notarikon
• Gematria

And the angel of the Lord said to him: Why have you struck your donkey these three times? Behold, I have come out as an adversary, for the road has been turned aside before me.

— Numbers 22:32

If one wrote a single letter as a notarikon, Rabbi Yehoshua ben Beteira deems him liable, and the Sages exempt him. Rabbi Yohanan said in the name of Rabbi Yosei ben Zimra: From where in the Torah do we learn the language of notarikon? As it is said: “For I have made you the father of a multitude of nations.” Av — I have made you a father to the nations; bahur — I have made you chosen among the nations; hamon chaviv — I have made you beloved among the nations; melekh — I have made you a king among the nations; vatik — I have made you venerable among the nations; ne’eman — I have made you trustworthy among the nations. Rabbi Yohanan himself said: Anokhi is a notarikon for ana nafshi ketavit yahavit — “I Myself wrote and gave [it].” The Rabbis said: amira ne’ima ketiva yahiva — “a pleasant utterance, given in writing.” Some say that Anokhi read backward means: yahiva ketiva ne’emanim amireha — “it was given in writing; its sayings are faithful.” The school of Rabbi Natan said: “for the way was yarat before me” means: yare’ah, ra’atah, natatah — “it saw, it perceived, it turned aside.” The school of Rabbi Ishmael taught: karmel means “a full ear.” Rav Aha bar Yaakov said: “and he cursed me with a nimretzet curse” — a notarikon: “he is an adulterer, he is a Moabite, he is a murderer, he is an enemy, he is an abomination.” Rav Nahman bar Yitzhak said: “What shall we say, and how shall we justify ourselves?” — “we are upright, we are righteous, we are pure, we are contrite, we are holy.”

— Babylonian Talmud, Shabbat 105a

A. Is Notarikon Remez or Drash?

Introduction

In this essay we will discuss notarikon. This is not an ordinary hermeneutical principle, even though it appears in the baraita of the thirty-two hermeneutical principles of Rabbi Eliezer son of Rabbi Yosei the Galilean, where it is principle 30. It seems closer to the plane of remez than to that of drash. As we shall see, one of the sources cited in the Gemara for notarikon comes from our weekly portion, from the verse: “for the way was yarat before me.”

Writing on the Sabbath

The talmudic passage cited above deals with the laws of writing on the Sabbath. As is well known, in order to incur the Torah-level liability of stoning or a sin-offering, one must write at least two letters. The question is what law applies to someone who wrote a single letter that functions as a notarikon. Rashi explains there that this refers to someone who wrote a dot above that letter to teach us that it is to be understood as a complete word; today we usually indicate this with a superscript. Since it carries the meaning of a complete word, is he liable as though he had written two letters or not? Rabbi Yehoshua ben Beteira and the Sages dispute this in the Mishnah there, Shabbat 104b, and the halakha follows the Sages, who exempt him.

By contrast, there is the well-known ruling, also cited there in the discussion on Shabbat 104b, that if someone writes one letter that completes a scroll, he is liable by Torah law, because in that case even one letter has significance. If so, it is not clear why, in the case of one letter that is a notarikon, we do not also follow significance. After all, that one letter marks a complete word. Why does this not give it sufficient significance to incur liability for writing it?

To understand this, we must first understand the basis of the prohibited labor of writing and its minimum measure, as well as the concept of notarikon and its meaning. What is notarikon at all? In its usual sense, notarikon means an acronym or acrostic. In the continuation of the talmudic passage, several sources are brought for notarikon from the Torah. Let us begin by trying to understand the movement and structure of that discussion.

First Source: “Abraham” or “Father of a Multitude”

The first source brought in the Gemara, by Rabbi Yohanan in the name of Rabbi Yosei ben Zimra, is Abraham as a notarikon for “father of a multitude of nations.” It is commonly understood as the initials of the words av hamon goyim, and this is indeed how Rashi explains it on Genesis 17:5. Rashi there also explains the apparently superfluous resh in the word Avraham.

According to this interpretation, something interesting emerges. In this case, we are not the ones expounding a word in the Torah by means of notarikon, as is usually the case in the rabbinic corpus. Rather, the Torah itself is performing the notarikon, that is, it itself writes both the word and its full explanation. We may note that we saw a similar situation in the essay on Parashat Beha’alotekha, where we pointed out that with regard to two principles we find the Torah itself explicitly using them: kal va-homer (a fortiori reasoning), and “two verses that contradict one another.”

This interpretation fits very well with the opening line of the Gemara: “From where in the Torah do we learn the language of notarikon?” According to Rashi, we have found a source in the Torah that one may use notarikon. The Torah itself uses notarikon.

However, this interpretation is difficult. First, as we have already mentioned, the letter resh seems to have no meaning here. Beyond that, the word “nations” does not appear in the notarikon at all. So this is, at best, only a rough approximation to a notarikon.

Moreover, the Gemara itself proves that this is not what it means by notarikon. For the Gemara brings a list of words that expand the notarikon, and these are not the words “father of a multitude of nations”: av, bahur, hamon chaviv, melekh, vatik, ne’eman. These words expand the sequence av hamon, and all the qualities hinted at here are said of him in relation to the nations. That is the meaning of the phrase “father of a multitude of nations” in the verse. So the initials here are the words av hamon, not the word Avraham.

But this explanation of the Gemara is also difficult, because it provides no source in the Torah for notarikon. At most, it offers a drash, or perhaps only a remez, on words in the Torah by means of the method of notarikon. If so, it is still unclear from where we know that one may do this at all. In addition, the similarity in the verse between Avraham and av hamon goyim hints that there really is a notarikon in the Torah itself, and that it is precisely the unfolding of the word Avraham.

All these difficulties with the Gemara, of course, do not resolve Rashi’s interpretation, which here takes a different path from the Gemara in understanding the biblical verse as notarikon.

The Other Sources: “Anokhi,” “Yarat,” and Others

As noted, Rabbi Yohanan is the sage who brought the previous source. Yet he himself apparently is not comfortable with it. Immediately afterward in the discussion, he brings an alternative source, this time from the Ten Commandments: Anokhi is a notarikon for ana nafshi ketavit yahavit. It is not clear why Rabbi Yohanan himself does not accept Rabbi Yosei ben Zimra’s exposition. It is possible that he thinks like Rashi on the Torah, namely, that the words in the verse are themselves the expansion of the notarikon in the word Avraham. But if so, he should have brought the word Avraham as his own example of notarikon, rather than going all the way to the Ten Commandments.

Other sages derive different notarikons from the word Anokhi. The Rabbis say: amira ne’ima ketiva yahiva. And there are those who derive a backward notarikon: yahiva ketiva ne’emanim amireha. All these are expositions on the Torah, teaching that the Holy One Himself gave it to us. Here too it is unclear why all of them specifically choose the word Anokhi, out of all the words in the Torah. Put differently: why must this word be expounded at all? And why specifically through notarikon? Rashi addresses this and explains that the exposition is made because the verse says Anokhi rather than ani. But that still does not explain why the exposition must be by means of notarikon.

The third source is brought by the school of Rabbi Natan from the word yarat in our weekly portion. The unfolding of the notarikon is: “it saw, it perceived, it turned aside.” Several additional sources of the same kind appear later in the discussion, and we need not spell them all out here.

All these sources are expositions of Torah words by way of notarikon. It is not at all clear from where the Sages derive the words that expand the notarikon. Is this free creation, without any rules? Do we find such interpretations in halakha (Jewish law) as well? Should we take such an exposition seriously, or are these only word-games played by preachers before an audience?

Beyond all this, all these sources face the same difficulty already noted above: at the opening of the discussion, the Gemara was seeking a source for the very legitimacy of expounding by notarikon. But none of these gives such a source. The discussion brings only examples of such expositions made by the Sages. The only apparent exception is Abraham, since there at least there is a biblical hint to a notarikon performed by the Torah itself. Yet, as we saw, even in that case the Gemara seems to ignore this and bring another notarikon instead. It seems that we must seek the solution precisely in that exposition.

The Source from Abraham Our Patriarch

It is hard to ignore the fact that in the case of Abraham our patriarch, the verse itself performs a notarikon. Therefore it may be that the midrash is to be read as follows: from the verse itself we learn that one may employ notarikon. Once we have learned that, we ourselves apply notarikon and expound that God promises Abraham all those promises through the words av hamon.

That is why Rabbi Yohanan himself then applies notarikon to the word Anokhi. He is not disputing Rabbi Yosei ben Zimra. Rather, he learns from him that one may use the method of notarikon independently, and he immediately applies it himself — “Rabbi Yohanan himself,” in the language of the Gemara — to the Ten Commandments.

Resolving All the Difficulties

According to this, all the difficulties are resolved. The first source really is a source from the verse that one expounds by notarikon. There is no dispute at all between that source and the later opinions in the discussion. Those later opinions are not trying to provide a source for the principle of notarikon, because in fact they provide no such source. As we have seen, they simply use this principle to expound various verses. According to our approach, all those sages are employing notarikon only after learning its legitimacy from the verse about Abraham. That source is accepted by all.

This also resolves Rashi, who explained that we learn notarikon from the verse about Abraham itself. We noted that this seemed, at first glance, to contradict the Gemara. According to our approach, this is in fact exactly what we find in the Gemara as well — after our reconstruction, which is somewhat like saying that something is missing from the text and must be supplied. The Gemara asks: from where do we know notarikon? The implied answer is: from the verse about Abraham. This is such a clear source that the Gemara does not bother stating it explicitly, and instead uses the method of notarikon both on that verse and on others.

The Principle of Notarikon

But we still have to understand what “notarikon” means. As we have seen, this is not a matter of initials in the full and precise sense. It is true that with regard to the missing resh in Avraham, Rashi goes to some trouble to explain it, and this implies that in his view notarikon really ought to be initials. But we have already pointed out the absence of the gimel from the word “nations.” Likewise, in the word yarat, the tet is not the first letter of the word hidden in it, namely natatah.

From here it clearly follows that notarikon is not necessarily a strict acronym. It is a compressed hint to hidden content, related to initials but not bound to be exact initials. This can be seen even more sharply in another midrash, in Mekhilta de-Rabbi Ishmael, Yitro, tractate de-BaHodesh, parashah 8, on the words “Honor your father”:¹

“Honor your father and your mother.” If you honor them, your days will be lengthened; if not, your days will be shortened. For the words of Torah are notarikon — that is how the words of Torah are expounded: from the affirmative one hears the negative, and from the negative one hears the affirmative.

Here we see that notarikon is virtually a synonym for remez in general. Whenever the Torah encodes information in some compressed way, this is called notarikon. See on this Torah Temimah, Genesis 17:5. Even an inference of the form “from the affirmative you infer the negative” is called notarikon.

It is true that in Rashi we saw that he does try to preserve the notion of initials, but that is only because the letter resh in Avraham seems to have no justification at all if this is indeed a notarikon. In the case of the tet in yarat, however, there is some justification for the letter, and therefore Rashi sees no need to explain it even though the initials are not exact. We may note that Rashi himself, on Parashat Yitro, brings the term “notarikon” regarding the inference “from the affirmative you hear the negative,” which means that he too accepts that notarikon need not be literal initials.

Is There Notarikon in Halakha?

At first glance, the method of notarikon looks somewhat questionable. Is it really a genuine hermeneutical principle, or perhaps only remez? Can it be applied in halakha as well?

As we already mentioned, this method appears as principle 30 in the baraita of the thirty-two hermeneutical principles of Rabbi Eliezer son of Rabbi Yosei the Galilean. If so, it is indeed a genuine hermeneutical principle. On the other hand, several sources state explicitly that this principle is not used in halakha. For example, from Ozar Midrashim, Eisenstein edition, “Thirty-two Principles in Aggadah,” p. 269:

“(30) Notarikon. From where do we know that one expounds notarikon in aggadah? Scripture says: Abraham — father of a multitude of nations; karmel — a full ear, something tender and crushed by hand; ‘and he cursed me with a nimretzet curse’ — initials: adulterer, Moabite, murderer, enemy, abomination; yarat — initials: it saw, it perceived, it turned aside.”

Here we see explicitly that this is an aggadic hermeneutical principle. Rashi writes similarly on Genesis 30:8:

“Naftulei — my prayer was accepted. And there are many aggadic expositions in the language of notarikon.”

As already noted, the thirty-two hermeneutical principles appear in Sefer Keritut. Rabbi Samson of Chinon goes through each one and explains why it is not included among the thirteen hermeneutical principles of Rabbi Ishmael. Regarding notarikon he writes as follows, in Sefer Keritut, Netivot Olam, principle 30:

“This too is aggadic exposition, and not words of Torah law, nor legal judgment. Therefore Rabbi Ishmael did not count it.”

The Criterion for Applying Notarikon

The criterion for applying this principle is not clear. Which words should be interpreted by means of notarikon, and which should not? What are the rules for choosing the words hidden in the notarikon? The unfoldings presented in the Gemara look rather wild. Some are even in Aramaic, and it is hard to suppose that this is a serious attempt to decode something genuinely hidden in the Torah itself.

It may be that some of these expositions really are no more than homiletical play. But it is difficult to assume that they are all of that sort, since this is a hermeneutical principle like the others in Rabbi Eliezer’s list. On the other hand, the criteria are not clear, and therefore it is quite understandable that this principle would be difficult to apply in halakhic exposition.

Apparently, wherever we encounter a word that has no accepted meaning, like yarat,¹ or a word that seems superfluous, like Anokhi instead of ani, as Rashi explains in the Gemara, we look for an interpretation by way of notarikon.

Back to the Prohibition of Writing on the Sabbath

As we noted at the outset, the Sages exempt one who writes a single-letter notarikon. The question is why they exempt him, when a single letter that has significance in other circumstances — for example, when it completes a scroll — does carry the Torah-level liability of stoning or a sin-offering. It should be noted that this ruling does not appear in the Tur or the Shulchan Arukh, but it is ruled as halakha by Maimonides, Mishneh Torah, Laws of Sabbath 11:13.

Perhaps the root of the matter lies in the question of what notarikon really is. If that one letter genuinely pointed to a complete word, then a complete word has been written. But if notarikon is only a kind of remez and not drash, then the word is not actually hidden inside the letter that marks it. According to our approach, the word is only hinted at by this letter, and the reader is the one who adds it himself.

In our essays on Parashat Chayei Sarah, Yitro, Pekudei, and elsewhere, we pointed out that according to Maimonides, all the laws derived through the thirteen hermeneutical principles and through inclusionary derivation — see his comments in the Second Root — are not contained within the scriptural text itself. The reader or interpreter expands the text by means of the hermeneutical principles. According to Maimonides, only the peshat (plain meaning) is present in the text, and only peshat is a tool of disclosure, whereas drash is a tool of expansion, not of disclosure. According to other commentators, all the hermeneutical principles are disclosive rather than expansive. That is, the laws derived by means of these principles are indeed already present in the text, and the methods of drash merely reveal them.

It may be that in the case of notarikon, because of the vagueness and flexibility of this principle, everyone agrees that the hidden word is not really contained in the letter that marks it. Notarikon, for everyone, is an expansive rather than a disclosive principle. If so, then even when we write a word as a notarikon by means of a single letter, we are not placing the word inside the letter; we are only hinting at it. The “unfolding” of the notarikon is not a disclosure of what was hidden within it, but rather an expansion added by the reader.

Accordingly, one who writes a word as a single-letter notarikon has not written the word itself, but only a letter, and the reader is the one who turns that letter into a full word. Therefore we should not be surprised that one who writes a single-letter notarikon on the Sabbath is not liable for it. It is possible that this itself is the point of dispute among the tannaim in the Mishnah: whether notarikon is remez or drash.

It may be that in a case where one letter constitutes an objective notarikon, universally recognized by all, the person who writes it would indeed be liable to stoning or a sin-offering even according to halakha. For example, one who writes a character in Chinese script, where a single sign has the meaning of a whole word — that is, a complete unit of meaning — would be liable.

A Concluding Reflection

We have seen that the principle of notarikon is very vague. It is not at all clear how and where it should be applied. Beyond that, it is not clear whether it is really a full hermeneutical principle, or only remez. After all, gematrias are merely side dishes to wisdom, and are generally treated as remez rather than drash. Why, then, should notarikon be any different? It should be noted, however, that gematria too is listed in the baraita of Rabbi Eliezer son of Rabbi Yosei the Galilean, as principle 29. If so, it would seem that within the hermeneutical principles of aggadah, remez is included as well.

We mentioned the words of the author of Sefer Keritut, who explains with regard to each of the thirty-two principles why it is not included among the thirteen principles of Rabbi Ishmael. As we mentioned in the essay on Parashat Vayeshev, the author of Sefer Keritut offers three different kinds of explanation, each relevant to different principles. One of these explanations is that there are principles that belong only to aggadic exposition, and therefore they are not included in Rabbi Ishmael’s baraita. But why exactly are these principles relevant only to aggadic midrash and not to halakhic exposition? On that, Sefer Keritut offers no answer.

From our discussion here it seems that at least part of the difference between the hermeneutical principles of aggadah and those of halakhic interpretation lies in their degree of precision and unambiguity.² The principles of aggadic exposition allow the expositor greater freedom, and less tightly dictate the result of the exposition. In any case, this matter still requires further study.

B. The Holy Tongue

Introduction

We have seen that one who writes on the Sabbath is liable only if he writes at least two letters. A single letter does not suffice — and even in the case of notarikon we saw that there is a dispute.

At the outset, it is important to note a fundamental difference between letters and words: a letter by itself is an arbitrary sign with no meaning. The basic unit of meaning in language is the word. If so, we would have expected the minimum measure for writing on the Sabbath to be a word, not two letters. We must also ask why the measure for writing on the Sabbath is defined in terms of the number of letters rather than the number of words.

The notarikon, with which we dealt in the previous section, is a method that transfers a word into a letter. A single letter represents a complete word, and thus meaning is stored within one letter. This is something like an inversion of letter into word, or word into letter. We pour meaning into a linguistic sign that by its nature is arbitrary.

Now, with regard to the Holy Tongue — that is, the language of the Bible — there are several indications that letters themselves, and not only words, carry meaning. In Sefer Yetzirah, as in other rabbinic sources, the letters are conceived as the building blocks of the universe. Such a conception assumes that the letters themselves also have meaning in some sense.

We find additional halakhic contexts in which letters, not words, appear. For example, in Babylonian Talmud, Shabbat 116a, it is explained that if a Torah scroll has been erased and eighty-five letters remain in it — the number of letters in the section beginning “And it came to pass when the ark traveled” — it still retains sanctity, and one saves it from a fire on the Sabbath. According to one of the amoraic opinions there, those letters may be “scattered,” that is, isolated letters rather than letters forming words. If so, it appears that a Torah scroll from which only scattered letters remain, in sufficient number, still retains something meaningful. Once again, the measure is letters rather than words.³

It is quite clear that this feature — granting meaning to letters — exists only in the Holy Tongue, and not in all languages. As we have seen, in other languages letters are merely arbitrary signs that have no meaning in themselves.

A similar phenomenon may be observed even with respect to words. In fact, words too — whether written or spoken — are arbitrary signs. True, unlike letters, words do have meaning; nevertheless, the choice of a particular word to designate a particular content is certainly arbitrary. One could have chosen any word to designate any content, so long as that choice were accepted by all the speakers in the community. In the next section we shall note that in the Holy Tongue, the situation is apparently different.

The Holy Tongue⁴

Several medieval authorities discuss the question of why the Holy Tongue is holy. Maimonides, in the Guide of the Perplexed — see part 2, chapter 30, and part 3, chapter 8 — explains this by saying that no names were coined in it for the sexual organs. The Nazir explains there that Maimonides is consistent with his own general view, since he understands the Holy Tongue as a conventional language like all other languages.

Several medieval authorities — see Nahmanides on the Torah at the beginning of Parashat Ki-Tissa, and Abraham ben David’s commentary to Sefer Yetzirah 1:10 — attack Maimonides’ approach and argue against him that such a language should have been called pure or clean, but not holy. On holiness and impurity, see the essay on Parashat Chukat. Therefore they argue — and so too does the Kuzari, 2:67-80 and 4:25, among others — that the Holy Tongue is called so because it is the language suited to the nature of beings and to true divine knowledge.

The meaning of this is that in the Holy Tongue, words and letters are not arbitrary. Their nature corresponds to the nature of the world itself. Therefore the description of the world and of divine truths in those words is the most perfect possible. In other words, the Holy Tongue is not conventional like all other languages. It belongs essentially to the nature of reality. Hence every translation into another language is only an interpretation, and will necessarily be less faithful than the original in the Holy Tongue.

The Nazir points out that a deaf-mute is regarded as lacking sound understanding because inability to express oneself in words reflects a deficiency in thought itself. And indeed Maimonides, in his Commentary on the Mishnah, explains this in a more rationalistic way, naturally consistent with his own view. There he holds that a deaf-mute is considered lacking sound understanding because his inability to speak and hear technically prevents him from learning wisdom.

Halakhic Implications

The Nazir brings two halakhic implications of this dispute. The first concerns the laws of reciting the Shema. The Mishnah states — see the beginning of the chapter “These Are Recited” in tractate Sotah — that the Shema may be recited in any language, not only in the Holy Tongue. Maimonides rules this in Laws of Recitation of the Shema 2:10, and adds: “One must articulate carefully in that language just as one articulates carefully in the Holy Tongue.” Abraham ben David, in his glosses there, is astonished by Maimonides’ words, and says that all other languages are only interpretation. How, then, can one — or should one — be exact about the letters of a text that is only an interpretation? Clearly both follow here their respective positions concerning the nature of the Holy Tongue. We have already seen that Abraham ben David, in his commentary to Sefer Yetzirah, disputes Maimonides on this point.

Another example is the reading of the Megillah. Maimonides, in Laws of Megillah 2:4 — and see Maggid Mishneh there — writes that if a person knows a foreign language and also knows Hebrew, he may read the Megillah in either of the two languages. By contrast, Nahmanides, in his novellae at the beginning of chapter 2 of tractate Megillah, holds that only if that person does not know Hebrew at all may he read it in the vernacular. Here too they follow their broader positions concerning the essential equality or inequality of languages.

Back to the Labor of Writing on the Sabbath

There is a puzzling ruling brought by Rabbi Moshe Isserles in Orach Chayim 306:11:

It is permitted to buy a house in the Land of Israel from a non-Jew on the Sabbath, and one may sign and register it in their courts. Gloss: in their courts, in their script; for this is prohibited only by rabbinic law, and for the sake of settling the Land of Israel they did not impose the decree.

Rabbi Isserles cites this in the name of the Or Zarua, who explained the permission mentioned in the Gemara in tractate Gittin to buy a house in the Land of Israel on the Sabbath and write a deed for it. According to him, all this applies only to writing in another language, because writing in such a language is prohibited only by rabbinic law. But in the Holy Tongue it is forbidden, since writing in it on the Sabbath is prohibited by Torah law.

Admittedly, this is a puzzling view and the subject of major dispute — and in practice we do not rule in accordance with it — but one can nevertheless see here a fundamental distinction between languages with regard to the prohibition of writing on the Sabbath. According to the Or Zarua, only writing in the Holy Tongue is prohibited by Torah law. Why should there be such a difference?

It seems that according to those who hold this view, the labor of writing is prohibited only when the script essentially represents the content itself. As we have seen, such a thing occurs only in the Holy Tongue. This may be the reason for the surprising ruling of the Or Zarua.

If that is indeed the definition of the labor of writing, then we may perhaps also understand why its minimum measure is stated in letters rather than words. The reason would be that in the Holy Tongue, the basic unit is the letter, not the word. The essential reflection of content depends on letters, not necessarily on words.

Perhaps this is also why notarikon is not prohibited by Torah law. True, the notarikon symbolizes a complete word, but that one letter does not reflect content of sufficient significance in the Holy Tongue — after all, the minimum measure is two letters. The word symbolized by it is only the meaning attached to the letter, not content essentially expressed through it. Therefore that letter is not considered a word.

Again: Writing a Letter That Completes a Scroll

In light of what we have said, the law of one who writes a single letter that completes an entire scroll requires further discussion. Why is that single letter considered to represent content of sufficient significance to incur liability? In fact, one may ask even more broadly: why is there a minimum measure of two letters at all? What is present in two letters that is absent from one?

It seems that the labor of writing requires an essential representation of reality, but also requires a minimum quantity of this phenomenon in order for one to incur liability, just as all Torah measures assign quantity to qualities.⁵ The reason is not that two letters reflect content and one letter does not. As we have seen, even one letter reflects content, but quantitatively it is not significant enough to incur liability.

When a letter completes an entire scroll, it completes the representation effected by the whole scroll. It turns everything that preceded it into a representation of the content expressed by the scroll. Therefore in that case the single letter is regarded as full of content that is also quantitatively sufficient. But a letter written as a notarikon tries to complete its measure by means of a word that is only its symbol. Yet symbolic relation is characteristic of foreign languages in their relation to the contents expressed in them, and therefore, according to the Or Zarua, this is not considered essential writing.

Of course, all this can be said only according to the view of the Or Zarua. In practical halakha, we accept that writing in any language is prohibited by Torah law, and that the measure is always two letters. According to those views, we would need a different kind of explanation, one that is not connected to the issue of language and its nature. But our concern here is not the labor of writing on the Sabbath in itself, only the law of writing on the Sabbath as an example that reflects some of the ideas we have discussed.

The Name of God⁶

A similar phenomenon to the one described here may also be found with regard to the names of God. One may refer to a human being, or to any other object, in two different ways: by description or by name. Names are arbitrary words attached to persons. A name may be chosen arbitrarily, according to the parents’ wishes. A description, by contrast, is not arbitrary. For example, if we say that a certain person is fat, we are not free to say that he is thin. The description depends on reality and is therefore not arbitrary.⁷

In the introduction to the book Shenei Luchot HaBerit, in the section called Beit Hashem, Rabbi Isaiah Horowitz notes, in the name of Rabbi Moshe Cordovero, that the situation with regard to the name of God is not like this. We have two modes of addressing the Holy One: names and appellations. The names of God are essentially connected to the very being they represent. By contrast, an appellation is a description of some divine attribute or manifestation. It follows, then, that the name of God relates to His essence in the way that a description of a human being relates to that person.

A question now arises: what is the difference between a divine name and an appellation? After all, an appellation too is a kind of description. At first glance, it seems that both forms of referring to God correspond to only one of the forms of referring to a human being.

It therefore seems that when we say that the name of God is an essential representation of God, we do not mean a description of Him. We mean an essential reference to His essence. Every description expresses one of the features of the person or object to which it refers. A name, by contrast, refers to the essence, not to any particular property.

In the case of a human being, the name is attached arbitrarily. In the case of the Holy One, however, His names belong to His essence in an essential way. That is why the names of God are found in the world of Atzilut — in Kabbalah, “names” and “sefirot” are almost synonymous terms. They emanate from God Himself, and therefore they do not merely “describe” Him; rather, they are, as it were, connected to His hidden essence.

This is not the place to enter into the meaning of this picture. Our aim here has only been to illustrate the meaning of essential attachment, as opposed to accidental description. The relation between the name of God and His essence parallels the relation between a word in the Holy Tongue and the reality designated by it.⁸ By contrast, the relation between a human being’s name and that person’s essence parallels the relation between a word in other languages and the content designated by it.

An Apology

Toward the end of our remarks we have departed somewhat from our usual custom and drifted into elevated matters. This is not the place to spell them out, elaborate upon them, and justify them, and we are certainly not free of error. Whatever mistakes we have made are ours to bear, and may we be cleansed of those hidden from us….

Footnotes