1

Published in B. Isaac, Y. Shahar (eds)

Judaea-Palaestina, Babylon and Rome: Jews in Antiquity

(Tübingen, 2012)

How do you say haroset in Greek?

Susan Weingarten

The spring festival of Passover commemorates the exodus of the

Jewish people from Egypt. When the Temple stood, it was the occasion

of a pilgrimage to Jerusalem, followed by the offering and eating of the

paschal lamb, together with matzah [unleavened bread] , and bitter herbs.

After the destruction of the Second Temple in 70CE, the rabbis recreated

the celebration of Passover with a ritual which takes place in Jewish

homes around the dining table, consisting of reading and discussing

texts, and pointing out and eating symbolic foods, as well as a festive

meal. This is conducted according to the seder [order], written in the

collection of texts known as the haggadah [narrative].

One of the special foods of the seder is haroset, a concoction now

usually made of fruit and spices, although its composition has varied

considerably over time and place. In this paper I want to look at the

origins of haroset .

Haroset in the Mishnah

Passover is mentioned in the Bible, together with the eating of

lamb, unleavened bread [matzah] and bitter herbs [maror], but not

2

haroset.1 What appears to be the first written evidence of haroset under

its own name is to be found in the Mishnah. However, this only mentions

haroset as such, without specifying its function, symbolism, ingredients

or taste.

They bring before [the leader of the Seder] unleavened bread [matzah]

and lettuce and the haroset, although haroset is not a religious

obligation [mitzvah]. R Elazar ben Zadoq says: It is a religious

obligation.2

Here the Mishnah mentions the unleavened bread [matzah], the bitter

herbs and haroset as belonging to the Seder ritual ‘even though haroset is

not a religious obligation,’ merely, it is implied, a custom. However the

Mishnah then quotes the contrary view of Rabbi Elazar ben Zadoq who

says that haroset is a religious obligation.

Rabbi Elazar, according to Tosefta Betzah iii 8, was a merchant in

Jerusalem who lived at the time of the destruction of the Temple. The

Babylonian Talmud quotes him as saying that merchants would cry the

spices for haroset in the streets of Jerusalem, where Jews would come on

Passover to eat the paschal lamb in the days when the Second Temple

still stood: ‘Come, buy your spices for the mitzvah [religious

requirement] [of haroset].’ 3 The Jerusalem Talmud earlier had cited the

‘merchants [tagrei] of Jerusalem’ crying their spices. If Jews were

coming to the Temple in Jerusalem for Passover, this would sound, then,

as if Jews must have already been eating haroset there together with their

roast lamb and unleavened bread and bitter herbs before the Temple was

1 Exodus 12:8

2Mishnah Pesahim x, 3

3BTPesahim 116a; JTPesahim 37d and parallels. I have translated this as ‘buy,’ since I consider this is

the meaning of the text, which actually writes ‘take.’ However there are those who would disagree:

Rabbi David, the grandson of the Rambam, for example, thinks the merchants were offering spices

free!

3

destroyed. However, Tosefta Pesahim x 10 talks of Rabbi Elazar and the

merchants of Lydda/Lod, rather than Jerusalem, so he might actually have

been talking about the time after the destruction of the Temple, not

before.4 In that case, the fact that the Mishnah reports a discussion as to

whether haroset is a mitzvah or not, might just mean that it was

something new, which belonged to the rabbis’ re-creation of the Passover

rituals as the Seder held in every Jewish home, rather than something that

went back to Passover in Jerusalem before the Temple was destroyed.

Note, however, from Rabbi Elazar’s evidence about buying spices

for haroset, whether it was in Jerusalem in Temple times, or in Lod after

the Destruction, that it is clear that Rabbi Elazar certainly saw spices as

an essential ingredient of haroset.

The Mishnah continues, in words quoted at every Passover Seder

by the youngest child present:

On all other nights we dip our food once, on this night we dip twice.5

The Mishnah is talking here first of all of the usual practice all over the

Roman empire of dipping bread into a condiment at a meal (among the

poorest this was sometimes all the meal).6 This happened every day. But

at the Seder, it says, we dip twice. Nowadays this refers to an initial

dipping of ordinary herbs into salt water and the later dipping of the bitter

4 In Tosefta Pesahim x,10 Rabbi Elazar is quoted as saying to some merchants of Lydda ‘Come, buy

your spices for the mitzvah [religious requirement] [of haroset].’ See on this: S Friedman Tosefta

Atiqta (Ramat Gan 2002, in Hebrew) 421-438. Friedman thinks that in this case the Tosefta ante-dates

the Mishnah, and as in many other cases, the commoner ‘Jerusalem’ was substituted for Lydda.

Friedman proposes that Tos Pesahim must be referring to a time after the destruction, when the rabbis

assembled in Lydda, rather than Jerusalem, as recorded in Tos Pesahim iii 11. However, Friedman does

not discuss the evidence of Tos Beitzah that R Elazar was ‘a merchant in Jerusalem all the days of his

life.’ 5 MPesahim x, 4. This is quoted in the Haggadah as part of the ‘Four Questions’ asked by the youngest

child. 6 Cf Ruth 2:14

4

herbs into haroset. In earlier times the bitter herbs were probably ritually

dipped into haroset twice at the Seder, just as there are two ritual hand-

washings.7 This custom seems to have persisted in some places until the

early Middle Ages, when Rabbi Isaac ben Moses of Vienna disapproved:

eating haroset before the right time (ie before the second dipping with the

bitter herbs), he said, was like sleeping with one’s fiancée before the

wedding.8 The bitter herbs themselves are specified in Mishnah Pesahim

ii.6:

And these are the herbs by [eating] which at Passover a man fulfils

the mitzvah: hazeret, olshin, tamkha, harhavina and maror.

The Jerusalem Talmud clarifies the meaning of two of these terms for us:

hazeret is hasa, ie lettuce, while olshin is translated by the Greek word

entubin, which means endives, or troximon, which refers to the raw

vegetable salad which accompanied a meal. The other terms are less clear

and will not concern us here. Lettuce was clearly bitter in those days, as

sometimes even today.

The discussions in the Mishnah, the Tosefta and the Jerusalem

Talmud discussing dipping hazeret are abbreviated in the extreme. They

do not actually say in so many words that this should be dipped into

haroset, but this can be inferred from the discussion in JT Pesahim x 3;

37c-d, and the later more explicit discussion in BT Pesahim 115a.

But there is no discussion at all in the Mishnah of what haroset is. People

were obviously expected to know. Even if its presence at the Seder was

7 S Friedman Tosefta Atiqta (Ramat Gan 2002, in Hebrew) 421-438; J. Kulp The Schechter Haggadah

(Jerusalem, 2009) 183-188. Probably the ordinary herbs were first dipped into haroset too, instead of

salt water, as today. On the history of dipping foods at the Seder, see in particular J. Tabory ‘The

history of the first dipping on Passover eve in the period of the Mishnah and the Talmud’ Bar Ilan 14-

15 (1977) 70-78 (in Hebrew); and in English id. ‘Towards a history of the Paschal meal’ in PF

Bradshaw, LA Hoffman (eds) Passover and Easter: origin and history to modern times (Notre Dame

1999) pp.62-80; and now J. Tabory JPS Commentary on the Haggadah: historical introduction,

translation and commentary (Philadelphia, 2008) 23-4 8 Rabbi Isaac ben Moses Or Zaru’a (ed A Marienberg, Jerusalem, 2006).

5

new, this was clearly a case of making a new use of what must have been

a known food.

An earlier hint of haroset?

Paradoxically, it is not the Mishnah, a Jewish text, which has the

first possible allusion to haroset, but a Christian one. We have just seen

that the first Jewish mention of haroset probably came in the third-

century Mishnah, but this may be predated by the New Testament, for the

famous Last Supper eaten by Jesus and his disciples before the

crucifixion was almost certainly the Passover meal.

The gospel of Matthew is considered by scholars to be the closest

of all the gospels to Jewish roots. It was probably written towards the end

of the first century. It writes:

And the disciples did as Jesus had appointed them; and they made ready

the Passover.

Now when even was come he sat down with the twelve.

And as they did eat he said, Verily I say unto you, that one of you shall

betray me.

And they were exceedingly sorrowful, and began every one of them to

say unto him, lord, is it I?

And he answered and said, He that dippeth his hand with me in the dish,

the same shall betray me.9

Only three verses after this does the gospel say that Jesus ‘took bread and

blessed it and broke it,’ so in talking about dipping in a dish the text is

unlikely to be referring here to the common ancient practice of dipping

bread in a condiment. This seems to be a separate dipping. Could the

writer of the gospel be referring to dipping bitter herbs into haroset here?

9 Matthew 26: 19-23. There is a parallel in the gospel of Mark, generally dated around the same time or

slightly later. I am grateful to Professor Adele Reinhartz for discussion of the dating of the gospels.

6

As already noted, it is not clear from the Mishnah whether haroset

was being used for dipping already before the destruction of the Temple,

or whether it began when the rabbis re-created the Passover Seder. If

Jesus really did dip bitter herbs into haroset this would bring us back to

his lifetime, the time when the Temple still stood. However, here too the

evidence is unclear. The gospel of Matthew was written some time after

the destruction of the Temple, and we cannot know whether the author is

using a genuine tradition of what really happened at the Last Supper

while the Temple still stood, or whether he is depicting it in terms of the

Seder meal he knew from his own times. But if the tradition were

genuine, this could be the earliest allusion we have to dipping the bitter

herbs in the haroset, paradoxically preserved for us in a Christian text.

There is further Christian evidence about bitter herbs at the Last

Supper, this time from the church father Jerome, writing in Bethlehem in

the fourth century, around the time of the compiling of the Jerusalem

Talmud. Jerome was in contact with many Jews and often shows that he

knew their customs well.10

He translated the gospels and much of the

Bible from Greek and Hebrew into Latin. The Greek word used by the

New Testament for the ‘dish’ Jesus and Judas dip into is tryblion, a bowl.

In his Latin translation of the gospel of Matthew, Jerome translates this

Greek word with another Greek word (!),paropsis. Could there be a

particular reason for this? The term paropsis is usually used to mean a

little side-dish holding appetizers to be eaten before a meal, but the word

could also be used for the food itself, rather than the dish. It was

especially used to refer to bitter foods like edible bulbs, bolboi, and

10

S Weingarten The Saint’s Saints: Hagiography and geography in Jerome (Leiden, 2005), passim

7

stalks.11

So Jerome is using a word here which implies the use of bitter

herbs. Is this because of his knowledge of the practices at the time of the

New Testament, or is he writing from his knowledge of what Jews did at

Passover in his own contemporary Palestine? Sadly, here too there is not

enough evidence to be certain.

We now turn back to our Jewish sources, where both the Talmuds

expand the discussion of haroset which we saw in the Mishnah, giving us

more details.

The Jerusalem Talmud

The Jerusalem Talmud expands and comments on the Mishnah:

… They bring before [the leader of the Seder] unleavened bread

[matzah] and lettuce and the haroset …

The people of the house of Issi [said] in the name of Issi: And why is

[haroset] called by the name of dukkeh? Because she pounds [dakhah]

[it] with [him/them/it] [corrupt text].

Rabbi Joshua ben Levi said: It must be thick. That is, in memory of clay.

Another tanna teaches: It must be thin. That is, in memory of blood.12

11

A Dalby Food in the ancient world from A to Z (London, 2003) 118; 212 who quotes Athenaeus

ii63d-64f. See also Deip ix 367c-368c. GB Dannell “Samian cups and their uses”, in Romanitas:

Essays on Roman archaeology in honour of Sheppard Frere on the occasion of his ninetieth birthday,

ed RJA Wilson, (Oxford, 2006) 147-176 notes the Samian ware dipping bowls from La Graufesenque

from the time of Nero which were labelled: paropsis/ides trocliati/bol[. It is quite possible that these

descriptions refer to the contents, the root troc- referring to the Greek troximon, raw vegetables and

bol[ from bolboi, bitter bulbs, both often eaten as appetizers to be dipped in sauce. We have already

noted that the olshin [endives] used as bitter herbs was translated by troximon by the Jerusalem

Talmud, and bulbusin [=bolboi] also appear in the Rehov mosaic inscription (l.4) among other produce

of the Sabbatical year. See: Y Feliks Masekhet Shevi’it II (Jerusalem, 1986)451-3, who also identifies

them in JT Demai 22c. Bolboi are still eaten in rural Greece and Italy:AF Buccini’The bitter – and

flatulent – aphrodisiac: synchrony and diachrony of the culinary use of Muscari Comosum in Greece

and Italy’ in SR Friedland (ed.) Vegetables: Proceeding of the Oxford Symposium on Food and

Cookery 2008 (Totnes, 2009) 46-55. 12

JTPesahim x, 3; 37c-d

8

The Jerusalem Talmud here quotes Issi (probably a rabbi, although it does

not say so) who notes that haroset is called ‘dukkeh,’ and asks why. The

answer given is that this is because it is pounded [dukhah]. 13

The Hebrew name dukkeh for haroset has survived to the present

day. Jews from the Yemen, cut off for many centuries from the

mainstream Jewish community, relied on the Jerusalem Talmud as their

religious authority, unlike other Jews, for the Babylonian Talmud did not

reach them for many hundreds of years. The Yemenite Jews have

preserved the tradition of the Jerusalem Talmud, and to this day the

Yemenite Jewish community in Israel still calls haroset ‘dukkeh.’14

We

may also note here the use of the name dukkeh among Palestinian Arabs

for a condiment made of pounded hyssop (za’atar) and sesame seeds.15

The Jerusalem Talmud then goes on to quote Rabbi Joshua ben

Levi, a third-century rabbi from the Land of Israel, as saying that haroset

must be thick, like mud or clay. This clay was what the Children of Israel

used for making bricks when they were slaves to Pharoah in Egypt, as we

read in the biblical book of Exodus. However, another rabbi, the

anonymous tanna, says that haroset should be liquid, ‘in memory of the

blood.’ Blood from the Passover lamb was used by the Jews to mark their

houses before they left Egypt. They dipped bunches of hyssop into blood

and painted it on the lintels of their houses as a sign for the Destroyer to

pass over them and spare their children. However, the mention of blood

may also refer to the first of the ten plagues in the book of Exodus, when

all the water in Egypt was turned to blood.16

The Jerusalem Talmud thus

13

The text is corrupt here but this is the reading proposed by Lieberman Yerushalmi KiFshuto ad loc. 14

Y. Kappah Sefer Agaddata dePasha (Jerusalem, 1958). 15

C Roden The New Book of Middle Eastern Food (Harmondsworth, 2000) p00000.I am grateful to

Youval Rotman for reminding me of this condiment. 16

Blood on lintels: Exodus 12. 21-3; plague: 7.20f

9

has two competing symbolisms for haroset: clay and blood, both

memories of different aspects of the Egyptian slavery.17

The Jerusalem Talmud also quotes the Mishnah about R Elazar ben

Zadoq noted above,which shows that it was usual to put spices in haroset.

However, there is no mention of any other ingredients, and no details of

taste.

The Babylonian Talmud

In the Babylonian Talmud, finalised around the seventh century,

there is a discussion of haroset in tractate Pesahim which deals with

Passover:

Rav Pappa said: …This lettuce, you must plunge it in the haroset,

because of qappa... [but others disagree, saying] you do not need to

plunge it in, because qappa dies from the smell…

Rav Pappa also said: A person should not leave the bitter herbs [maror]

in the haroset, just in case the sweetness of the spice destroys its

bitterness…18

Rav Pappa, a Babylonian rabbi of the fourth century, tells us that haroset

is used as a dip for the lettuce used as bitter herbs, to counteract the

harmful qappa they contain. It is unclear here from the text what qappa

is: – unhealthy juices or some form of worm are the alternatives

proposed by mediaeval commentators, but it may also have been an evil

17

These were finally reconciled in the Middle Ages by the Tosafist R Elijah miLondres Etz Hayyim (ed

MYL Sacks with Cecil Roth, Jerusalem, 1956) p153 18

BTPesahim 114a-116a

10

spirit. Elsewhere in the discussion we are told of other possible antidotes

to qappa in different vegetables:

the qappa of lettuce is counteracted by radishes, the qappa of radishes

by leeks, the qappa of leeks by hot water; the qappa of all these by hot

water.

This makes qappa sound like harmful juices. However, the text of the

BT adds a spell you can say if you have no antidote to the qappa, which

makes it sound more like a demon or an evil spirit:

Let him say thus: qappa, qappa, I remember you and your seven

daughters and eight daughters-in-law.

Later qabbalistic rabbis gave all sorts of symbolic meanings to the

mother with her seven daughters and eight daughters-in-law, but these

need not detain us here.19

We should just note that the Babylonian

Talmud is very often particularly concerned with demons and spirits, so

this may be a later Babylonian addition, rather than part of earlier

concepts.20

Rav Pappa, then, thinks that the bitter herbs need to be dipped in

the haroset in order to counteract the qappa, although other rabbis think

this unnecessary. Rav Pappa adds that the bitter herbs should not,

however, be left in the haroset, since this would destroy their bitterness.

Presumably this is the origin of the custom among present-day Jews

originating from Germany not to eat their haroset at all: they just dip in

the bitter herbs and then eat them, having shaken off any haroset

adhering to them.21

19

Eg Vilna Ga’on Be’urei Aggadot on BT Berakhot 57b, (repr. Jerusalem, 1971) 44-45. 20

See now G Bohak Jewish Magic (Cambridge, 2008) 21

CF Tabory, Haggadah p104.

11

The rabbis then discuss how many portions of matzah, bitter herbs

and haroset were needed, and rule that only one was necessary, for the

leader of the Seder. They then carry on to discuss once again the question

of whether haroset is a religious obligation [mitzvah], bringing the

Mishnah we have already seen with Rabbi Elazar ben Zadoq’s evidence.

The discussion then continues, quoting Rabbi Levi and Rabbi Yohanan,

and later Abbaye, asking now what haroset signifies:

What is the mitzvah?

Rabbi Levi says: In memory of the apple.

But Rabbi Yohanan says: In memory of the clay.

Abbaye says: Therefore you have to make it acidic and to make it thick.

… 22

The text here records the discussion of the multiple symbolic meanings of

haroset, which serve to remind the participants in the Seder of various

aspects of the deliverance of the Jews from slavery. Thus Rabbi Levi

proposes that haroset is ‘in memory of the apple.’ There are a number of

Midrashim and commentaries which make it clear that this refers to

rabbinic exegesis on the apple-tree in the biblical Song of Songs (8.5).23

Rabbi Levi, we should note, does not actually say that apples should be

included in the haroset, but it is clear that, for him, apples are closely

connected to haroset and its symbolism.

Rabbi Yohanan, on the other hand, thinks that haroset is a memory

of the clay for the bricks which the Jews made as slaves in Egypt. We

22

The printed versions of the Talmud have here: Make it acidic, in memory of the apple. And you must

make it thick in memory of the clay. This seems to be a comment by Rashi which got interpolated in

the text, as it is missing in the manuscripts. See Diqduqei Soferim ad loc. 23

Midrash Exodus Rabbah 1.12.4 with parallels in Land of Israel midrashim and Babylonian Talmud

Sotah 11b, etc.

12

have already seen that the Jerusalem Talmud also presents haroset as a

memory of the clay.

We should note here that both Rabbi Yohanan and Rabbi Levi are

rabbis from Palestine, so the Babylonian Talmud here appears to be

reporting a debate which actually took place in Palestine, but was not

preserved in Palestinian sources.

In order to reconcile these differing opinions, the Babylonian

Talmud carries on to cite Abbaye, a fourth-century Babylonian rabbi,

who says that haroset must be made both acidic and thick. Apples clearly

were more sour than sweet in late antique Babylonia. Indeed, maybe they

were more sour in antiquity in general than today. The first century

Roman writer Pliny writes of apples in his time in Italy, including the

wild apple ‘with a horrible sourness … so powerful it will blunt the edge

of a sword.’24

Thus apples can be used to make haroset acidic.

We now have a concluding piece of evidence that supports first

Rabbi Yohanan’s opinion, and then Rabbi Elazar ben Zadoq’s ruling that

haroset is a mitzvah:

There is a baraita like the statement of Rabbi Yohanan: The spice is in

memory of the straw, the haroset is in memory of the clay.

Rabbi Elazar ben Zadoq said: the merchants in Jerusalem used to say:

‘Come, take your spices for the mitzvah.’

Thus the Babylonian Talmud reports that the rabbis apparently decided

the debate in favour of Rabbi Yohanan: haroset is to be seen as a memory

of the clay, rather than a memory of the apples. Rabbi Yohanan is quoted

24

Pliny HN 15.52

13

again, saying that the spices in the haroset are in memory of the straw25

used for making bricks, while the haroset itself is in memory of the clay.

Thus the Babylonian Talmud associates haroset with apples, but

rejects the view that this is the major association, and does not actually

speak of apples as an ingredient, only an indication of taste. The

important thing is that haroset should be thick like clay, though it should

be acidic too, and that spices should be added, in memory of the straw.

Haroset – not only a Passover food

Is it possible to go any further in finding out about the early origins

of haroset from Jewish sources?

It is clear that haroset, although given special symbolic status at

the Passover Seder, was in fact eaten all the year round, not only on

Passover. We find instructions in the Mishnah that flour is not to be

added to haroset or to mustard on Passover, in case they ferment and

become hametz [leaven].26

If it was added by mistake, the haroset must

be eaten straight away, before it had time to ferment.

The Babylonian Talmud mentions a special vessel where the

haroset was prepared, called the beit haroset, although it is unclear

whether this existed in Palestine as well.27

Whereas in those days Jews

were allowed to use most everyday crockery on Passover if it had only

been used for cold food and was then cleaned, there were two vessels that

were forbidden, because they might be too difficult to clean – the vessel,

beit se’or, where sourdough was fermented in order to make leavened

25

Cf Exodus 5.6-18. 26

Mishnah Pesahim ii, 8 27

BTPesahim 30b

14

bread, and the beit haroset, forbidden ‘because it [the acidic haroset] also

ferments very strongly.’28

Seder and symposium

We have already discussed the earliest Jewish sources on haroset,

and found that these clearly pre-dated the redaction of the third-century

Mishnah, but that it was not totally clear whether haroset existed at the

time the Temple still stood, or whether it belonged to the re-creation of

Passover by the rabbis after the destruction of Jerusalem and the Temple.

So far our Jewish sources. We have seen that the New Testament only

hints at haroset, and does not name it.

What about other sources? Is there any evidence of haroset or

something like it in Greek or Latin literature? By the end of Temple

times, when we first hear of haroset, the Land of Israel had long been

under the influence of Rome. While rejecting the religious cults of the

Romans, Jews lived their lives in the Graeco-Roman world and willingly

or unwillingly took part in its culture. Their everyday language appears to

have been Aramaic, but some of them spoke Greek too, and many Greek

words (including food terms) found their way into the texts of the

Mishnah and the Talmud. King Herod rebuilt the Temple using Greek

orders of architecture. It is clear from evidence from the Tosefta that

some Jews in the Land of Israel dined like Greeks or Romans on

occasion.29

So we should not be too surprised to find that the Jewish

celebration of Passover, the Seder meal itself, has many similarities to

Graeco-Roman cultural practice. As already noted, once the Temple was

in ruins it was no longer possible for Jewish families to gather and eat

28

loc cit 29

Tosefta Berakhot iv 8

15

their roasted lambs ‘with matzah and bitter herbs’ in a mass out-door

picnic near the Temple. The rabbis thus re-instituted Passover, moving it

into the home of every Jew. Many of the elements of this new celebration

around the Seder table were shared with the Graeco-Roman symposium,

with its formal ceremonies.30

Both the Seder and the symposium were

meals where the participants reclined on couches, where discussion of

literature and philosophy and of the foods eaten formed part of the

conversation, where wine-pouring and wine-drinking played an important

part, and which often began with eating appetisers such as lettuce and

eggs.31

There is a third-century mosaic found in Antioch [Antakya, in

modern Turkey] with pictures of the courses of a meal, from beginning to

end: first a silver trulla [ladle] for hand-washing, then the first course on

a round flat plate: its eggs, artichokes and lettuce with a bowl of sauce to

dip them into, and flat bread at its side, which might have reminded us of

a Seder table – if it were not for the accompanying pigs’ feet.32

Graeco-Roman parallels?

Aside from the general similarities between Seder and symposium,

it has been suggested by a number of scholars that there may be a specific

connection between Graeco-Roman sauces and haroset. Stein looked in

Athenaeus’ Deipnosophistai, the Philosophers at Dinner, an account of a

Greek symposium from the third-century,33

contemporary with the

30

See on this S Stein ‘The influence of symposia literature on the literary form of the Pesah haggadah’

JJS 8, (1957) 13-44, contra: B Bokser The origins of the Seder: The Passover rite and early Rabbinic

Jusdaism (Berkeley, etc., 1984); PF Bradshaw, LA Hoffman (eds) Passover and Easter: origin and

history to modern times (Two liturgical traditions vol 5) Passover and Easter: origin and history to

modern times: the symbolic structuring of sacred seasons (Two liturgical traditions vol 6

(Indiana1999); D Smith From symposium to eucharist: the banquet in the early Christian world

Minneapolis 2002 31

On the symposium see in general A Dalby, Siren Feasts: A History of Food and Gastronomy in

Greece (London/NY, 1996 repr. 1997); on the Seder as symposium see n. 29 above. 32

F Cimok (ed.) Antioch mosaics (Istanbul, 1995) 47 33

Athenaeus came from Naucratis in Roman Egypt, but it is unclear where his book was written

16

Mishnah, and wrote that dishes ‘similar though not identical’ to haroset

‘are described at length by the same author.’34

Stein describes haroset as

being made of ‘nuts and fruits pounded together and mixed with spices,

wine or vinegar,’ and his references to Athenaeus are to the entry ‘nuts’

in the index of Gulick’s edition. However, attempts to identify haroset

solely on the basis of its modern ingredients are doomed to failure. The

ingredients, and hence the taste, have varied over time, and while almost

all haroset today is sweet, this was not so in the past, nor are nuts an

essential ingredient. The earliest hint we have of taste is in the

Babylonian Talmud’s citation of rabbis from Palestine seen above, which

notes the disagreement as to whether haroset should be thick, like clay, or

sharp like apples, presumably sour apples. Later Ashkenazi rabbis took

this definition ‘sharp’ quite literally: the 11th

century Rashi, for example,

identifies haroset as aigros, ‘sour’ in Old French.35

Nuts do not appear

Rashi’s haroset either, or in that of the 12th century Maimonides, and

appear to have been popularised only by the mediaeval Tosafists.

If ingredients alone are not the solution to haroset, then, we must

turn to philology and function.

What is the Greek for haroset?

The Cairo Genizah, an assembly of discarded Jewish manuscripts

from mediaeval Egypt, preserves a fragment of a Mishnaic glossary,

explaining words from the Mishnah by using Greek words transliterated

into Hebrew characters. This has been published by Nicholas de Lange,

who dates it to ‘earlier than the tenth century,’ making it one of the

earliest documents in the Genizah. This glossary includes the information

34

Stein art. cit. n. 23, 16 35

Rashi on BT Pesahim 30b

17

that haroset in Greek is tribou enbamous, written אנבמוסטריבו . 36 De

Lange identifies tribos as a path or a socket, and was unable to identify

enbamous at all. However, tribou would seem to come from the verb

tribo to pound or grind, whence the Greek term for a sauce, trimma.37

Archestratus of Gela, a fourth-century BCE food writer whose work is

preserved by Athenaeus, writes of a dipping sauce made by pounding

(tripsas). 38

Enbamous would appear to refer to the Greek word

embamma, which is used to mean a sauce used as a dip, deriving from the

verb embapto, embaptomai to dip. Later in the same passage of

Archestratus, the verb embapto is used for dipping into a pounded sauce.

In their commentary on this passage, the editors Olson and Sens describe

the verb embapto as ‘the vox propria for dipping food in a side-dish sauce

or the like.’ Thus Archestratus uses both terms found in the glossary as an

explanation of haroset in his instructions to dip (embapte) food into a

sauce made of pounded (tripsas) ingredients. The ingredients used by

Archestratus for his specific sauce are specified as hyssop and vinegar.39

We have already noted the presence of hyssop dipped in blood at the first

Passover in Egypt, as recorded in Exodus 12.22. Both hyssop and vinegar

are present (among other things) in the recipe for haroset given by

Maimonides in the 12th

century.40

Clearly, however, not all forms of

embamma were made with hyssop and/or vinegar. Xenophon notes many

different kinds among the Medes, while Pliny specifies a form of

embamma made with pounded mint.41

Interestingly, the haroset of

36

N de Lange Greek Jewish texts from the Cairo Genizah (Tübingen, 1996) p304 no.16, l. 12 37

HG Liddell,R Scott A Greek-English lexicon (Oxford, 1996, 9th

ed.) sv. Cf. Dalby Food in the

ancient world, n 11 above, p.294. 38

Archestratus of Gela: Greek culture and cuisine in the fourth century BCE eds S Douglas Olson and

A Sens (Oxford, 2000), Fragment 23 (Brandt 22) line 5, quoted by Athenaeus Deip vii 326 f. 39

. The Greek of Archestratus has hyssopos here, like the Septuagint text of Exodus 12.22, although

Olson and Sens prefer to translate it as marjoram, and Gulick, in his translation of Athenaeus Deip vii

326f, as caper-plant. On the confusions between these plants, see Dalby Food in the ancient world

above, n.11, sv Hyssop, and bibliography ad loc. 40

R Moses Maimonides [Rambam]: Commentary on the Mishnah: Pesahim 10, 3 ad loc 41

Xenophon Cyropaedia 1.3.4; Pliny HN 20.53. 147

18

Maimonides’ grandson, Rabbi David, has mint instead of hyssop,

together with vinegar, among its other ingredients.42

It is also noteworthy that embapto is the verb used by Jesus in the

two gospel accounts (Matthew 26.23 and Mark 14.20 ) mentioned above,

where Jesus talks of the disciple who dips (embapsas, embaptomenos)

with him in the dish.

Archestratus, we saw, was cited by Athenaeus, who was

contemporary with the Mishnah. For our next piece of evidence, we turn

to the one Graeco-Roman cookery book which has survived from

antiquity. This collection, de re coquinarea, 43

attributed to Apicius, in its

present form dates to fourth-century Rome, but it certainly includes

earlier information from the first and second centuries CE. The Apicius

collection often uses Greek words for cooking terminology, in the same

sort of way as French culinary terms are used by present-day English

cookery writers aspiring to haute cuisine. The collection includes two

ways of dressing lettuce and endives. The translation which follows is

cited from the new edition of Grainger and Grocock, slightly adapted:44

Endives and lettuces

3.18.1 Correct [the bitterness of ] endive with a dressing of liquamen, a

little oil and chopped onion. But instead of lettuce in winter serve endive

in enbamma45

or with honey and sharp vinegar.

Lettuces: [serve] with oxypor[i]um, with vinegar and a little liquamen46

for the digestion, and to ease wind and to prevent the lettuces from

doing harm:

42

Rabbi David b Abraham Maimuni, grandson of Rambam (1222-1300) also called R David HaNagid,

Midrash on the Passover Haggadah (tr. from the Arabic by S. Barh”i, Jerusalem, 1981, in Hebrew),

p37. 43

We shall distinguish Apicius the man from the collection bearing his name by italicising the latter:

Apicius 44

C Grocock, S Grainger (eds and trs) Apicius: a critical edition with an introduction and an Englis

translation of the Latin recipe text (Totnes, 2006) iii,18, 1-2. 45

MS. var: enbamma 46

Liquamen was the famous Roman salty fish sauce also known as garum. Liquamen, together with all

the other very popular Graeco-Roman fish condiments, disappeared from general use around the 10th

century. On the complexities of Roman fish condiments, including lengthy discussions of liquamen and

garum, see the comments by Sally Grainger in her edition of Apicius (n.44 above). I am grateful to

19

2 oz cumin, 1oz ginger, 1 oz green rue, 12 scruples (= ½ oz) juicy dates,

1 oz pepper, 9 oz honey, Ethiopian, Syrian or Libyan cumin. Pound the

cumin after you have steeped it in with vinegar. When it has dried, mix

all the ingredients with the honey. When required mix half a teaspoonful

with vinegar and a little liquamen or take half a teaspoon after dinner.

These Apicius recipes, under the heading of ‘endives and lettuces,’ note

here that these dressings, sauces or dips ‘correct’ [medere] endives. It is

specifically stated that lettuce and endives are interchangeable, and can

be served either with ‘enbamma,’ or honey and vinegar. We note the form

of the word used here in one of the manuscripts, enbamma, is more like

that in the Genizah glossary than the classical Greek form, embamma.

The digestive sauce called oxypor[i]um is also included under the

heading of preparations for endives and lettuce, and while enbamma is

said merely to ‘correct’ lettuce and endives, oxyporium is noted as

specifically preventing lettuce doing harm, just as we saw that the

Babylonian Talmud claimed that haroset counteracted the harmful qappa

in the bitter herbs. Talmudic bitter herbs, as we have noted, were also

usually lettuce or endives.

Greek and Roman doctors such as Galen and Anthimus usually

saw lettuce as beneficial – or certainly less harmful than other vegetables.

The philosopher Athenaeus, writing about the Greek symposium,

suggests that lettuce is an anti-aphrodisiac. It has been claimed that it was

because of these supposed anti-aphrodisiac properties that the rabbis

wanted to use lettuce for the Seder, where the riotous revelry after the

banquet common at Greek symposia was discouraged. 47

However, we

should note here that lettuce was commonly eaten as an appetizer at

Greek and Roman banquets without any apparent worry by the

participants about its possible effect on their sexual function afterwards.

Sally Grainger for allowing me to see a pre-publication version of this discussion, as well as for her

comments, and Andrew Dalby’s, on oxyporium. 47

Bokser (above n.29)

20

But unlike all these, both the Talmud and the Apicius collection

agree on the potential harmfulness (or at least the bitterness) of lettuce

and endives. Haroset and enbamma (and oxyporium) were explicitly

used to correct this potential harmfulness. Unfortunately we are not told

of the ingredients of enbamma. However, many of the ingredients

mentioned in the Apicius oxyporium sauce – dates, as well as ginger,

cumin, pepper, vinegar, honey – appear in various forms of haroset,

albeit at a later date. Dates, and date honey [silan], indeed, are ingredients

of the earliest forms of haroset for which we have recipes, from ge‘onic

times, the recipes of Amram and Sa’adiah Ga‘on.48

On the other hand, I

have never seen rue in a recipe for haroset, although herbs in general are

sometimes mentioned. It may be relevant here that rue is a particular

favourite with the Apicius collection, appearing in about 20% of his

recipes, which could reflect the editor’s own individual preference, rather

than common practice. Rue is also extremely bitter in taste, and there was

a trend of incorporating bitter herbs into haroset both in Ashkenaz and

Sepharad in the Middle Ages. Liquamen, the fish condiment that is also

very common in the Apicius collection, is clearly not part of the actual

oxyporium, but added later with the vinegar. Vinegar or sour wine was

often added to haroset from the time of the ge’onim onwards, giving it a

sweet-and-sour or even totally sour taste: we have already mentioned

Rashi’s aigros, sour haroset.

In addition to all this, we noted above that the Jerusalem Talmud

called haroset: ‘dukkeh,’ as being something that was pounded. The dip

48

Seder Rav Amram Gaon ed. D Goldschmidt (Jerusalem, 1971); Siddur R Saadja Gaon eds I

Davidson, S Assaf, BI Joel (Jerusalem 1941) 134; the combination of pepper, cumin and ginger

appears in the 12th

century Sefer haRoqe’ah.

On these and on haroset in general see my paper ‘Haroset’ in R Hoskings (ed.) Authenticity in the

Kitchen: Proceedings of the Oxford Symposium on Food and Cookery 2005 (Totnes, 2006) 414-426

21

described by Archestratus above was made of pounded (tripsas) hyssop

and vinegar, and the Genizah glossary includes tribou in its translation of

haroset. Unfortunately Apicius does not give details of how to make

enbamma. But the recipe for the parallel oxyporium also used to

counteract the harmful effects of lettuce includes the instruction to pound

at least some of the ingredients. Graeco-Roman literature contains other

recipes for oxyporium, recommended for their digestive properties.49

The

sixth-century Greek doctor Aetius has an oxyporium with dates, which,

like the Apicius oxyporium, is very similar to haroset.50

The talmudic evidence about lettuce and haroset which we have

discussed here comes from the Babylonian Talmud, which was further in

place and time from the Graeco-Roman world than the Jerusalem

Talmud, and belonged to a different culinary culture. However, the

Babylonian Talmud does sometimes preserve and cite older, original

material that has been lost from its Land of Israel counterpart. If this is

the case here, then it is probable that haroset owes its origins to the

Graeco-Roman sauces whose functions were to counteract the bitterness

and/or ill-effects of lettuce and endives. These sauces were at least

sometimes made, like haroset, by pounding some of the ingredients, and

some of these ingredients overlap with known ingredients of haroset,

albeit at later dates. The philological evidence of the Cairo Genizah

glossary confirms that some mediaeval Jews identified haroset with the

dipping sauce known to the Graeco-Roman world by the Greek name of

embamma, made by pounding (tribo). We know that bitter herbs were

eaten on Passover from biblical times. However, in the Graeco-Roman

49

There are different recipes for oxyporum in Columella de re rustica 12, 59,4-5. Galen also mentions

‘the black garos that is called oxyporon in Latin’ as a digestive: On compounding 12.637. Andrew

Dalby Food in the ancient world from A to Z (London, 2003) sv digestives writes that this is because

garum (or liquamen) was used as a liquid vehicle for this mixture 50

Aetius 3,92 has ginger, pepper, cumin , parsley, fat dates, rue, honey and vinegar.

22

period, bitter lettuce and endives appear to have been generally identified

as harmful, and in need of dipping in a sauce to correct them. Could this

have been the reason for the introduction of haroset into the Passover

seder?*

*Some of the material in this paper was presented at the Oxford

Symposium for Food and Cookery on Authenticity, and at the Hebrew

University of Jerusalem, Department of Classics Seminar on

Contextualisng Graeco-Roman Antiquity. I am grateful to the participants

for their enlightening discussions. The research was supported by a grant

from the German-Israeli Foundation (GIF).