Did Proverbs Copy an Egyptian Book?

"It can hardly be doubted that the author of Proverbs was acquainted with the Egyptian work and borrowed from it." — Miriam Lichtheim, Egyptologist, 1976¹

"But if Proverbs is the borrower here, the borrowing is not slavish but free and creative. Egyptian jewels, as at the Exodus, have been re-set to their advantage by Israelite workmen and put to finer use." — Derek Kidner, Tyndale Old Testament Commentaries, 1964²

Summary

In 1924 the Berlin Egyptologist Adolf Erman showed that a newly published Egyptian papyrus, the Instruction of Amenemope, runs parallel to a block of the Book of Proverbs.  A skeptic reads this as proof that part of the Bible was lifted from a pagan book.  Some defenders answer that the Egyptian copied the Hebrew, or that both used an older source.  This article tests each answer against the manuscripts, the Hebrew text and the parallels themselves, with the four tests for dependence set out in the parent article:  verbal match, unusual detail, date order and route of contact.

1. Proverbs 22:17-23:11 used Amenemope. D. C. Simpson's 1926 table, still printed in the standard anthology, finds an Egyptian parallel for 21 of the 24 verses of this block, and for 1 of the 46 verses that follow it in the same collection.³ The opening call to listen, the angry man, the boundary stone, the meal before a ruler, and riches that grow wings and fly to the sky all appear in both. Michael Fox's Anchor Yale commentary of 2009 calls the parallels "too numerous and too specific to be reasonably ascribed to anything other than literary dependence."⁴
2. The number thirty is the signature. Amenemope has thirty numbered chapters and ends, "See thou these thirty chapters."³ Proverbs 22:20 asks, "Have I not written thirty sayings for you". The written Hebrew word there, shilshom, "the day before yesterday," appears 14 times in the Hebrew Bible, and 13 of those times it is paired with "yesterday." Only in Proverbs 22:20 does it stand alone, which is why scribes, the Greek translators and modern commentators have all struggled with it. Read with different vowels, the same letters give "thirty."
3. The direction runs from Egypt to Israel. The oldest copies of Amenemope, an ostracon in Cairo and a papyrus in Stockholm, date from about the 21st or 22nd Dynasty, roughly 1000-900 BC, and they are school exercises, so the book was older still.⁵ The theories that the Egyptian copied a Hebrew book (R. O. Kevin in 1930, Étienne Drioton in 1959) needed Amenemope to be late. Kevin placed its author among the Jewish colonies in Egypt "after the exile."⁶ The manuscripts are several centuries too old for that, and the Egyptian text contains puns that only work in Egyptian.⁷
4. A common source is possible but has no evidence. Irene Grumach's lost "Ancient Teaching" behind both books has no manuscript. The conservative Egyptologist John Ruffle, who doubts direct copying, says of it that it "has no basis in fact."⁵
5. The strongest conservative dissent lowers the degree of copying, not its direction. Ruffle counts about eleven verses as real matches, stresses the jumbled order, and gives the verdict "not proven." His own model is an Egyptian scribe at Solomon's court working "from his memories of a text."⁵ That still has the material coming from Amenemope.
6. The order is mostly rearranged. Placed in Proverbs' order, the Egyptian passages agree with Amenemope's own order in 69 of 104 pairs, against about 52 expected by chance. That is a weak tendency, not a copy run straight through. Fox explains it as an editor who went through the Egyptian scroll in five passes, taking what he wanted each time.⁴
7. Proverbs did not take the theology. The Hebrew writer adds the reason for the sayings, "So that your confidence may be in the Lord," and gives the poor a divine advocate: "the Lord will plead their case."⁸ Neither line has an Egyptian parallel.
8. Borrowing fits what the book says about itself. The section is headed "the words of the wise," and 24:23 adds "These sayings also are from the wise."⁸ Proverbs also includes the words of Agur and of King Lemuel. The conservative commentators Derek Kidner and Bruce Waltke accept the Egyptian model and treat the section as Scripture. Inspiration has never required that a biblical writer invent every sentence.
9. The parallel does not date the section. Amenemope was written by about 950 BC at the latest and was still copied in Egyptian schools in the seventh and sixth centuries. So the borrowing fits Solomon's court, as Ruffle, Kitchen and Waltke hold, and it fits Hezekiah's, as Fox holds. The evidence does not choose between them.

Verdict:  the skeptic is right that Proverbs 22:17-23:11 depends on an Egyptian book, and the defenders who deny it, or reverse it, are wrong on the evidence.  The skeptic is wrong that this shows the section to be mere borrowed literature.  The Hebrew writer took about ten Egyptian chapters, cut and rearranged them, set them inside a book that names its sources as "the wise," and grounded them on trust in the LORD.

Part 1: How the parallel was found, and how the debate ran

From a purchase in 1888 to Erman's paper of 1924

The British Museum bought the papyrus now numbered BM 10474 in Egypt in 1888.  E. A. Wallis Budge, its keeper of Egyptian antiquities, described its contents in 1922 and published the whole text in facsimile in 1923.⁵  In his English edition of 1924 Budge noted parallels "in the Pentateuch, the Psalms, and the Proverbs of Solomon."  He thought the influence might run the other way:  "it is impossible not to wonder whether these at least are not due to some Asiatic influence which found its way into Egypt".⁹

Adolf Erman read a short paper to the Prussian Academy on 1 May 1924 that turned the question around.  He wrote that the Egyptian book "touches the Solomonic book of proverbs directly in some places, and touches it in such a way that, in my view, no chance agreement can be thought of."  He counted nine certain contacts between Amenemope and the whole Book of Proverbs, and found that seven of them fall inside 22:17-24:22, a block scholars already treated as a separate small book:  "so kann doch hier von Zufall nicht die Rede sein," so chance cannot be the explanation here.¹⁰  His explanation of the route was a guess.  A Jew living in Egypt "in Saite or Persian times" learned the Egyptian school book, translated it into Hebrew or Aramaic, and "replaced 'the god' with his Yahweh."¹⁰

Erman's count can be priced.  The block 22:17-24:22 is 70 of the 915 verses of Proverbs, 7.7 percent of the book.  If nine contacts fell at random through Proverbs, the block would hold 0.7 of them on average.  The chance that seven or more of nine would land there is about 5 in 10 million.¹¹  The caveat is that the contacts were chosen by a scholar, not drawn at random, so the figure measures how concentrated Erman's own list was, not the whole case.

The flood of parallels, and the first dissent

Erman's paper set off what John Ruffle later called "an academic parlour game of "Spot the Parallel"" with "bonus points for emendations of the text."⁵  Hugo Gressmann claimed to find thirty sections in Proverbs 22:17-24:22 and to link each one to a chapter of Amenemope.  H. O. Lange published the standard edition in Copenhagen in 1925.  F. Ll. Griffith translated the text for the Journal of Egyptian Archaeology in 1926, and in the same issue D. C. Simpson printed the verse-by-verse comparison that John Wilson later reprinted in Ancient Near Eastern Texts.³

Simpson found the inference that almost all of 22:17-24:22 is "ultimately derived" from Amenemope, or from a source Amenemope also used, "would appear to be irresistible," and he dated the Hebrew section to the late monarchy.¹²  Others were less sure of the direction, and the reason matters.  In 1925 Lange had argued from the language that the book "kann wegen der Sprache nicht älter als die 21.-22. Dynastie sein," could not be older than the 21st or 22nd Dynasty, and that author and papyrus might both be as late as the Persian or Greek period.¹³  Griffith put the British Museum copy between the 25th Dynasty and the reign of Darius.¹⁴  If the Egyptian book was that late, it was younger than Solomon and perhaps younger than the prophets.

On that dating, W. O. E. Oesterley in 1927 proposed that both books drew on "an older Hebrew collection."  His reason was stated plainly:  "Assuming, as we do, that the Egyptian book was written after the high ethical teaching of the prophets had taken root and fructified among the finer spirits of their nation, it is a priori more likely that Amen-em-ope would be influenced by the Old Testament writers than vice versa."¹⁵  Robert Oliver Kevin of the University of Pennsylvania went further in 1930.  He argued that Amenemope's "work was directly called into being when a copy of a little book of thirty Hebrew proverbs "The Words of the Wise" fell into his hands."⁶  He relied on Lange's late date, which he took to mean "nearer 300 B.C."⁶

The consensus, Drioton's challenge, and Ruffle

By 1951 Walter Baumgartner could report that "the theory that Amenemope is the original of Prov. xxii. 17-xxiii. 11 has now been generally accepted," and that the attempt to trace both books to an earlier Hebrew collection "has been given up."¹⁶  In 1957-1959 the French Egyptologist Étienne Drioton revived the reverse theory with philological arguments.  He concluded that Amenemope "est en réalité une traduction en égyptien d'un livre hébreu aujourd'hui perdu," is in reality an Egyptian translation of a Hebrew book now lost, which the compiler of Proverbs also knew "dans son texte hébreu original."¹⁷  R. J. Williams answered him in 1961, and Irene Grumach proposed a common Egyptian source in 1972.  In 1975 the Egyptologist John Ruffle gave the Tyndale Biblical Archaeology Lecture on the question and reopened the case against direct copying.⁵  The positions those scholars took are weighed in Part 5.

The arc of the debate is the first finding of this article.  Every serious argument for Hebrew priority was made while Amenemope was thought to be late.  Part 2 shows what the manuscripts found later did to that premise.

Part 2: The manuscripts and their dates

Eight witnesses, and how they are dated

Amenemope survives complete on one papyrus and in pieces on seven other objects.  The dates below are each scholar's, with the reason he gave.  Most rest on the style of the handwriting, which in this period can be fixed only within a century or two.

Two facts stand out.  The full papyrus, the copy Erman and Lange worked from, is one of the latest copies, perhaps from the seventh or sixth century BC.  And the earliest copies are school exercises.  The Turin board has a pupil's day-marks in its margin, "Day 4, Day 8, Day 10, Day 12 and Day 18," which Ruffle reads as "set portions for a pupil's study."⁵  A book copied as a classroom text had been written some time before.

When was it written?

Every scholar read for this article who wrote after the early copies were known places the composition between the 19th and 21st Dynasties, roughly 1300 to 950 BC, or sets a floor in that range.  Lange and Griffith, writing in 1925-1926 before those copies were known, had dated it later.  The reasons differ.

1. R. J. Williams, 1961: on the strength of the Cairo ostracon, "Since its use as a school text indicates that it had been current for some time, we may well regard the work as originating as early as the Nineteenth Dynasty."⁷
2. Miriam Lichtheim, 1976: "The composition of the work is now usually assigned to the Ramesside period, although all the manuscript copies that have reached us are of later date."¹
3. John Ruffle, 1977: the early copies "indicate a date for the composition before 1,000 B.C.," and "even an XVIIIth Dynasty date would be acceptable."⁵
4. Harold Washington, 1994: "The text can be dated securely to the late Ramesside period on the basis of four types of evidence: (1) paleographical data, (2) personal names, (3) loan words, and (4) chapter headings." Because the earliest pieces are student exercises, "we must allow some decades to have elapsed before Amenemope became a standard exercise text," which puts its writing in "the earlier twentieth dynasty, 1186-1069 BCE."¹⁹ The name of Amenemope's father, Kanakht, is "otherwise attested no later than the twentieth dynasty."¹⁹
5. Michael Fox, 2009: "Amenemope was probably composed in the late 20th or 21st dynasties (1188-945 B.C.E.), but it was still being copied as a school text at least as late as the 26th or 27th dynasty".⁴

Grumach, who moved the date later than most, still sets a floor of the twelfth century BC from the spelling of some foreign words.¹⁸

Fox adds a caution that cuts the other way:  "No copies can confidently be dated even to the 20th dynasty (ending in 1069 B.C.E.)."⁴  He does not mention the Cairo ostracon there.  The early end of the case rests on Černý's dating of that sherd, which is reported by Ruffle, Williams and Grumach but was never published by Černý himself, and on Peterson's dating of the Stockholm papyrus, which Fox reports more widely.  That is a real weakness.  It does not touch the main point:  even Fox, who doubts the early copies, puts the composition before 945 BC.

What the dates settle

Kevin's theory needed the Egyptian author to meet Hebrew proverbs among "the Jewish colonies that were established in Egypt after the exile," that is, after 586 BC.⁶  Drioton likewise pointed to the Jewish colonies of the Persian period.  The Cairo ostracon, on Černý's date, is about four hundred years older than that, and the composition is older still.  Williams drew the conclusion in 1961:  Drioton's appeal to the Jewish colonies at Elephantine "thus loses its cogency."⁷  Kidner, writing for a conservative evangelical series, put it more bluntly:  "The contention that Proverbs has made use of Amenemope (and not vice versa) has now been greatly strengthened, if not clinched, by chronological evidence."²

The dates also bear on Solomon, whom Kitchen places in the tenth century, "fl., c.950 BC."²⁰  On every date in the list above, Amenemope already existed in his day.  So no date for the Hebrew section, early or late, can make it older than the Egyptian book.

Part 3: The parallels, verse by verse

The table

Proverbs is quoted from the NET Bible, Amenemope from John Wilson's translation in Ancient Near Eastern Texts.  The pairings are Simpson's, as Wilson prints them;  the Amenemope locator is the page and line of the British Museum papyrus.³  The grade applies the parent article's first two tests to each pair:

1. Close: the same image or the same sequence of words, or a detail the subject does not force.
2. Partial: the same subject with some shared wording.
3. Theme only: a shared topic, no shared wording.
4. None: no Egyptian parallel.

The quotations are copied from the NET⁸ and from Wilson.³  Simpson also pairs Proverbs 24:11 with a line of Amenemope's chapter 8, which Wilson does not translate, and finds nothing for the other 45 verses of 23:12-24:22.³

What the grades add up to

Of the fifteen units in the table, seven are close, five partial, two theme only and two none (counting 22:24 and 22:25 separately).  The close pairs carry the case:  the opening call, the thirty, the angry man, the boundary stone twice, the meal before a ruler and the winged riches.  They draw on chapters 1, 6, 7, 9, 23 and 30, and the whole table draws on the introduction and nine of Amenemope's thirty chapters.

Scholars who have worked through the pairs land in the same range.  Ruffle, the most skeptical, accepts 22:17-18, 20 and 29, 23:1-5 and 10, which he counts as "11 verses of Hebrew text out of 24," and says this "does not seem to me to be wholly conclusive evidence of direct borrowing."⁵  Fox, the most confident, finds that "Out of twenty-four verses in Part IIIa, fifteen have one or more strong parallels in Amenemope, two have thematic but not verbal affinities, and only seven have none."⁴  Kidner, writing for a conservative series, says the points of contact "are too many and too close to be a matter of coincidence."²

The cluster

The strongest single argument is where the parallels sit.  Simpson's table finds a parallel for 21 of the 24 verses of 22:17-23:11 and for 1 of the 46 verses of 23:12-24:22, though both belong to one collection under one heading.  Shared wisdom would spread its likenesses evenly.  A Fisher exact test puts the chance of so lopsided a split, if the likenesses were spread at random over the 70 verses, at about 1 in 10 trillion.²²  The same caveat applies as to Erman's count:  the table is a scholar's list.  But the point does not depend on the precise number.  Fox puts it in words:  if the likenesses were only "similar literary responses to similar situations in similar cultures," that "does not explain why the Amenemope parallels are so concentrated in a short unit."⁴

The order

Ruffle's best objection is order.  Listing the Egyptian passages in the order they appear in Proverbs gives the sequence "2, 10, 1, 3, 6, 4, 11, 9, 5, 7, 8, 4, an order which cannot be explained on grounds of sense or style."⁵  Oesterley and Kevin made the same point in the 1920s.

The order can be measured.  Simpson's fifteen paralleled units, placed by the page and line of their Egyptian passages, form 105 pairs.  In 69 the two books agree in order, in 35 they disagree, and 1 is a tie.  Shuffled at random, the same passages agree in about 52 pairs on average, and 69 or more turns up about 4.5 percent of the time.  The longest run that keeps Amenemope's order is 7 of the 15 units, against an average of 5.5 at random, which is not significant.²³  So the Hebrew writer did not copy straight through, and Ruffle is right about that.  There is a modest pull toward the Egyptian order.

Fox explains both facts at once.  He traces the editor making "five passes or "sweeps"" through the Egyptian scroll, moving forward each time and then going back, and notes that "a scroll does not allow for easy skipping about but requires constant unrolling and re-rolling."⁴  Within some passes the sequence holds:  Proverbs 22:17-18 follows Amenemope 3.9-16 in sequence, and Proverbs 23:5 splices together lines from 9.19 and 10.4-5.⁴  This is how ancient scribes used sources.  Fox notes that Egyptian sages quoted older books by "rephrasings, abbreviations, or expansions of the original— just what we see in Proverbs' treatment of Amenemope."⁴

Grade on the four tests

1. Verbal match: passed, for about seven units.
2. Unusual detail: passed: the "thirty," the winged riches, the widow's or orphan's boundary.
3. Date order: passed: the Egyptian is older (Part 2).
4. Route of contact: plausible on any date for the Hebrew section (Part 6).

Grade:  dependence shown, Egyptian to Hebrew, for 22:17-23:11.  The dependence is selective and free.  For 23:12-24:22 the evidence shows at most the wider Egyptian tradition.

Part 4: "Thirty sayings" and the Hebrew of 22:20

Three readings of one word

The Leningrad Codex, the oldest complete Hebrew Bible, gives Proverbs 22:20 as הֲלֹא כָתַבְתִּי לְךָ followed by a word written one way and read another, then בְּמוֹעֵצֹת וָדָעַת, "with counsels and knowledge."²⁴

1. The written text (ketiv) is שלשום, shilshom, "the day before yesterday," and by extension "formerly."
2. The reading tradition (qere) is שָׁלִישִׁים, shalishim. A shalish is an officer; Fox notes the word means "a charioteer (e.g., Exod 14:7; 1 Kgs 9:22) or an adjutant" and "bears no special connotations of excellence." The King James Version's "excellent things" comes from this reading.⁴
3. The Greek translators read τρισσῶς, "three times": "καὶ σὺ δὲ ἀπόγραψαι αὐτὰ σεαυτῷ τρισσῶς," "and you, write them for yourself three times."²⁵ The Latin Vulgate also reads "triply."²

All three readings, and "thirty," שְׁלֹשִׁים, sheloshim, share the consonants ש-ל-ש-ם.  They differ in vowels and in the vowel letters ו and י.  The Greek shows that its translators saw the root for "three" in the word.

Why the written word is suspect

The written word can be tested against the rest of the Hebrew Bible.  שלשום occurs 14 times.  In 13 of them it stands in the fixed phrase "yesterday and the day before," with תמול or אתמול:  Genesis 31:2;  Deuteronomy 4:42 and 19:6;  Joshua 3:4, 4:18 and 20:5;  Ruth 2:11;  1 Samuel 10:11, 14:21 and 19:7;  2 Samuel 5:2;  2 Kings 13:5;  and 1 Chronicles 11:2.  Proverbs 22:20 is the only place it stands alone.²⁶  Fox draws the conclusion:  "The ketiv slswm "the day before yesterday" cannot be correct, because the word appears only as part of a fixed idiom".⁴  The qere fares no better, since a word for officers makes no sense with "counsels and knowledge."  The verse was a known problem before anyone had seen Amenemope.

The case for "thirty"

Erman argued that the word, left with its natural meaning, means nothing other than "thirty," and pointed to Amenemope's closing address to his son, "You have now seen these thirty chapters."¹⁰  Most modern translations follow.  Wilson's note in Ancient Near Eastern Texts reads, "In Prov. 22:20, the Hebrew is to be read: "Have I not written unto thee thirty (sayings)?""³  The NET prints "thirty," and Tremper Longman notes that the NRSV and NIV make "that emendation with confidence."²⁷  So does Kidner, who calls it "the most convincing, for the section can be divided into this round number of paragraphs."²

Two facts from before 1924 support the reading without depending on Amenemope at all.

1. Felix Perles proposed "thirty" in 1905-1906, "with no knowledge of Amenemope."⁴ He reached it from the Hebrew alone.
2. Older commentators already divided the section into about thirty units. Fox reports that the tenth-century Jewish scholar Sa'adia's comments "fall into twenty-eight segments," and that Franz Delitzsch in 1872 divided 22:17-24:22 into thirty-two groups, or thirty-one without the opening.⁴

The thirty of Proverbs is not thirty borrowings.  Fox:  "there are not thirty maxims drawn from Amenemope, but thirty in the entirety of Part III."⁴  The Hebrew writer took the Egyptian book's frame, a teaching in thirty units, and filled it with shorter Hebrew sayings, about ten of them from Amenemope and the rest from other sources.

The case against

Ruffle gives the strongest objections, and they deserve full weight.

1. Grammar. "a numeral in Hebrew can only stand by itself when the noun it qualifies is in close proximity and unambiguously linked." "Thirty" alone would need a word such as "sayings" to have dropped out.⁵
2. The count is soft. "Once we begin looking for 30 sections it becomes easy to find them, but it is arguable whether they are indisputably clear and obvious."⁵
3. Other solutions exist. He suggests keeping the written word and adding one before it to read "previously," or an unknown sense of the root. "I am not convinced that the reading 'Thirty' is the only solution or that it inevitably points to a link with the Egyptian text."⁵

Each objection is real, and none is decisive.  The grammar objection is answered by the context:  the next words, "with counsels and knowledge," and the whole collection supply what is counted.  The count is soft, but the pre-1924 counts of 28 and 31 show that it is close to thirty without anyone looking for thirty.  And Ruffle's "previously" requires adding a word that no manuscript has, while "thirty" requires changing only vowels.  Fox's summary is fair:  even if one finds "twenty-nine or thirty-one maxims, the proximity to thirty (which is the only explanation of slysym / slswm that makes good sense) is unlikely to be coincidental."⁴

Verdict on 22:20:  "thirty" is the best reading by a wide margin, and it is the best reading on the Hebrew evidence alone.  Amenemope explains why the number is there.

Part 5: Three explanations, each in its defenders' words

1. Proverbs used Amenemope

This is Erman's view and the majority view since.  Lichtheim:  "It can hardly be doubted that the author of Proverbs was acquainted with the Egyptian work and borrowed from it ... Ever since Adolf Erman pointed this out there has been a consensus among scholars on a literary relationship".¹  Harold Washington calls Amenemope "the structural model for the collection of thirty sayings known as the "Words of the Wise" (Prov 22:17-24:22), and ... the material source for nine of the first ten of these sayings."¹⁹  Fox holds that the editor "was using a text, a single scroll," because the parallels stop abruptly at 23:11:  "If he had been [working from memory], the parallels would be more evenly distributed in Part III."⁴

Conservative scholars hold it too.  Kidner accepts the borrowing outright.²  Bruce Waltke, in the New International Commentary, writes that "The structural model for this collection, "Do I not write for you thirty sayings?" derives from the last chapter of Amenemope (27:6): "Look to these thirty chapters.""²⁸

Within this view there is a range.  Erman thought of a translation chopped up by later collectors.  Baumgartner already called it "old-fashioned to suppose that the borrowing was a purely mechanical process;  the translator treated his original with considerable freedom, adapting it to suit the rhythm of Hebrew, the land of Palestine, and the religion of Israel."¹⁶  Ruffle's memory model, below, sits at the far end of the range.

2. Amenemope used a Hebrew book

Budge first raised the idea, Oesterley half-held it, and Kevin and Drioton argued it in full.  Kevin's reason was the order:  "Whereas the Hebrew runs consecutively, Amen-em-apt does not. How otherwise this phenomenon can be explained except on the theory that Amen-em-apt was dependent upon the Hebrew, appropriated such of its ideas as were suitable to him and expanded them in his own way, it is difficult to see."⁶  Drioton's was a set of places where the Egyptian seemed awkward, as though translated:  "Un faisceau d'indices convergents permet donc d'établir que la Sagesse d'Aménémopé est en réalité une traduction en égyptien d'un livre hébreu aujourd'hui perdu."¹⁷  Kidner states the theological pull of this view honestly:  that Amenemope "sometimes rises to heights that are worthier of an Israelite and a Christian than of a polytheist ... creates an initial presumption that he is the borrower."²

Four facts defeat it.

1. The dates. The theory needs Amenemope after the exile or in the Persian period. The earliest copies are about four centuries older (Part 2).
2. The puns. Williams found word-plays "at least three times in Amenemope, involving words which are not amenable to word-plays in Semitic," which "is strong support for an original Egyptian text."⁷ A translator cannot preserve puns that do not exist in his source language.
3. The loanwords. Drioton's Semitic words in Amenemope are normal in Egyptian texts of the New Kingdom. Williams: "The presence of numerous Semitic loanwords in Egyptian texts of this period is evidence of extensive cultural contact, not of translation from Semitic originals."⁷
4. The Egyptian setting. Lange noted in 1925 that "both form and content are genuinely Egyptian," citing the chapter on the river ferry.¹³ Amenemope's author describes himself as an "Overseer of the Soil" who "establishes landmarks at the boundaries of the arable land," which explains why his boundary chapter is the longest.³ And Kidner notes that "it is the Hebrew text that tends to be clarified when it is read by the side of its longer Egyptian counterpart."²

Kevin's order argument cuts both ways:  a Hebrew writer selecting from a longer book will also scatter his sources.  Williams judged that Kevin's "superficial knowledge of Egyptian rendered his arguments valueless."⁷  Fox's verdict on the whole school is that it "was based on misunderstandings of the Hebrew and the Egyptian."⁴

3. Both used a common source

This view comes in three forms.

1. A lost Hebrew collection. Oesterley: "the suggestion that both Amen-em-ope and the compiler of Proverbs made use of an older Hebrew collection deserves more serious consideration than has yet been given to it."¹⁵ It rested on his assumption that Amenemope came after the prophets, which the manuscripts have removed.
2. A lost Egyptian book. Irene Grumach proposed in 1972 an "Alte Lehre," an ancient teaching of the late 18th Dynasty from which Amenemope expanded his chapters and which reached Proverbs through a Canaanite version. Ruffle, who wanted to weaken the case for direct dependence, still rejected it: "It is impossible to rule out such a possibility but it has no basis in fact and until some manuscript evidence for this text is discovered it can only be theoretical."⁵ Fox adds that Proverbs shows the influence of "passages from the beginning, middle, and end of Amenemope," and that four "better than" sayings outside this section (15:16, 15:17, 16:8 and 17:1) also trace to Amenemope, so the Hebrew sages knew more than a short core.⁴
3. A shared international stock. Simpson spoke of a common stock of wisdom without national boundaries, and Krispenz in 1989 and R. N. Whybray in 1994 argued, in Fox's summary, that the parallels are "no more than similar literary responses to similar situations in similar cultures."⁴ Ruffle made the same point memorably, by showing that the Aztec huehuetlatolli recorded by Sahagún in the sixteenth century echo Ptahhotep and Amenemope: "such precepts and images are universally acceptable and hence ... similar passages may occur in Proverbs and Amenemope simply by coincidence."⁵ This is true of the theme-only and partial pairs. It does not explain the cluster in 22:17-23:11, the thirty, or the winged riches.

Kenneth Kitchen, often named with this camp, does not argue for a common source in the work read for this article.  In 1977 he left the Amenemope question to Ruffle, and he described Solomon's book as "incorporating some data from other sages (22: 17ff.; 24:23ff.)."²⁰

Ruffle's dissent, at full strength

Ruffle is the best conservative witness against direct copying, and his case is more than a list of differences.  He shows that several proposed parallels concern different subjects:  23:6-8 is about a stingy host where Amenemope speaks of a poor man's property, and 23:9 is about a fool where Amenemope speaks of discretion.  He shows that Amenemope's favourite themes, the "heated man" and the "silent man," hardly appear in Proverbs.  He notes that Proverbs 22:26-27 has no Egyptian source, and that if a chapter had dropped out of Amenemope "we shall end up with 31 chapters!"  He also warns that some of the matches exist only after scholars emended the Hebrew to fit the Egyptian:  "It seems to be a case of wanting to have one's cake and eat it."⁵  His verdict:  "I believe it can merit no more definite verdict than 'not proven' and that it certainly does not exist to the extent that is often assumed".⁵

His own explanation is that the passage "was contributed by an Egyptian scribe working at the court of Solomon based on his memories of a text that he had heard and, may be, used in his scribal training. I believe this proposal fits all the requirements: a striking metaphor and the thirty chapter framework are remembered clearly, and the subject of much of the Teaching is recalled though the details and order are muddled."⁵

That is a theory of dependence on Amenemope.  It differs from the majority only on how the Egyptian text reached the Hebrew page.  Fox's reply is that it "comes close to the standard theory of dependency, except that it interposes the unnecessary intermediary of memory in a literate scribal culture."⁴  The evidence of Part 3, a modest pull toward the Egyptian order, fits both a scribe working from a scroll in passes and a well-trained scribe working from memory.  The evidence does not settle which.  It does settle the direction.

Part 6: Egyptian wisdom and the road to Jerusalem

Amenemope in its own tradition

Amenemope was the last of a long line.  The Instruction of Ptahhotep, which Lichtheim dates to "the latter part of the Sixth Dynasty," in the later third millennium BC, already has thirty-seven maxims that "do not amount to a comprehensive moral code, nor are they strung together in any logical order," much like Proverbs 10-29.²¹  Its seventh maxim gives the table manners of Proverbs 23:1:  "If you are one among guests / At the table of one greater than you, / Take what he gives as it is set before you; / Look at what is before you".²¹  The Instruction of Any, which Lichtheim dates "almost certainly in the Eighteenth Dynasty," warns against "a woman who is a stranger":  "A deep water whose course is unknown, / Such is a woman away from her husband."¹  Proverbs 23:27 calls the harlot "a narrow well."

These older books matter in two ways.  First, they show that many of the likenesses between Proverbs and Egypt are stock wisdom, which is why the grades in Part 3 separate the close pairs from the rest.  Second, they show how Egyptian scribes treated sources.  Fox:  "Amenemope himself used earlier texts, including Ptahhotep, Mentuhotep, the Ramesside Prayer to Thoth (AEL 2.114), and especially Anii".⁴  The Hebrew writer worked the way Egyptian sages had worked for a thousand years.

They also sharpen the one unusual detail.  Kitchen notes that Amenemope "has 30 numbered chapter-headings, a practice found almost 1000 years later in the 25 numbered 'teachings' in the Demotic Papyrus Insinger".²⁰  Washington:  "Amenemope is the only text of the Instruction genre to contain chapter divisions with the numbered superscription."¹⁹  A Hebrew book that speaks of "thirty" sayings is pointing at the one Egyptian instruction known to count its chapters.

The Proverbs collection also drew on non-Egyptian sources.  Fox identifies the Aramaic sayings of Ahiqar behind 23:13-14, on the rod that will not kill a child, and Any, or a book quoting him, behind the drinking poem of 23:29-35.⁴  The title "words of the wise" fits a collection with several sources.

How could a Hebrew writer have met it?

This is the fourth test, and it is where the two sides differ most.

The Solomonic route.  First Kings says Solomon married Pharaoh's daughter (3:1) and "was wiser than all the men of the east and all the sages of Egypt" (4:30).⁸  His court lists contain a puzzle.  David's scribe is called Seraiah in 2 Samuel 8:17, Sheva in 20:25, and Shavsha in 1 Chronicles 18:16, and Solomon's two scribes are "the sons of Shisha" in 1 Kings 4:3.⁸  Tryggve Mettinger concluded that "The forms Shisha etc. in the lists of high officials are due to a misunderstanding of the title" of an Egyptian royal secretary, "borrowed from Egypt," and that "No definite answer can be given as to whether this official was a native Israelite or an Egyptian."²⁹  If he is right, the office that wrote down the king's words used an Egyptian title.  That is the route Ruffle proposes:  "what could be more reasonable than that Solomon should question a cultured Egyptian at his court about this wisdom?"⁵

The Hezekian route.  Fox rejects the Solomonic setting because it "is based almost entirely on the biblical narrative ... whose reliability, especially for the semi-legendary era of the United Monarchy, cannot be taken for granted," and because, following Bernd Schipper, Egypt had little involvement in Israel and Judah between the end of its empire and the late eighth century.⁴  He argues that "it is virtually impossible that a Judean would have learned to read Egyptian, which requires years of schooling in the difficult hieratic cursive," and that "The most likely scenario for the translation and transmission of Amenemope is the early seventh century, and the most likely language for the translation was Aramaic."⁴  He grants that "Hieratic numerals were used by Israelite scribes in the seventh-sixth centuries, but there is no evidence for any broader knowledge of Egyptian language and writing."⁴

Both routes are possible, and each has a weakness.  The Solomonic route rests on the account in Kings and on Mettinger's reading of a scribe's name.  The Hezekian route rests on Fox's distrust of that account, and on an Aramaic translation of Amenemope that no one has found.  Fox's own route also assumes an intermediary, a translation, which is no stronger than Ruffle's intermediary, a memory.  A skeptic should notice that the best mainstream reconstruction puts the transmission in the century of King Hezekiah, whose "men" Proverbs 25:1 names as copyists of Solomon's proverbs.⁸  On either route the test of contact is passed.

Part 7: What Proverbs changed

Amenemope is not a crude book.  Lichtheim praises its "quality of inwardness," and says that in it "worldly success, which had meant so much in the past, has receded into the background. Even poverty is no longer viewed as a misfortune."¹  The Hebrew writer did not need to moralize it.  What he changed was the God behind the morals.

1. He named the LORD as the purpose. Proverbs 22:19, להיות ביהוה מבטחך, "So that your confidence may be in the Lord," stands where Amenemope's introduction says its teaching will "direct him to the ways of life."⁸ Fox counts 22:19 among the verses the editor "composed ... independently of his source."⁴
2. He gave the poor a divine advocate. Amenemope's second chapter forbids robbing the oppressed. Proverbs keeps the command and adds a verse with no Egyptian parallel: "for the Lord will plead their case and will rob the life of those who are robbing them" (22:23).
3. He replaced a vague threat with a redeemer. Amenemope warns the man who moves a widow's boundary, "Lest a terror carry thee off."³ Proverbs says of the orphans' fields, "their Protector is strong; he will plead their case against you" (23:11). Ruffle notes the difference: the Egyptian threat "is hardly comparable with the suggestion that the perpetrator will be disagreeably surprised to find that the defenceless orphan in fact has a powerful ally who will take him to court."⁵ The Hebrew is גֹאֲלָם, "their go'el,"²⁴ the kinsman-redeemer of Israel's family law.
4. He dropped the Egyptian gods. Amenemope's chapter on riches says "there is no ignoring Fate and Fortune," whom Wilson identifies as "The god Shay and the goddess Renenut," sends ill-gotten wealth sinking "into the underworld," and ends with a prayer "to the Aton when he rises."³ Proverbs 23:4-5 keeps only the image of riches flying off like a bird. Ruffle notes that the underworld lines are exactly what "a Hebrew writer might omit as unsuitable or incomprehensible."⁵
5. He gave moral reasons where the Egyptian gave social ones. On eating with a ruler, Ruffle observes: "Ptahhotep advised his reader how to behave generally to create a good impression, Amenemope wants him to avoid appearing greedy. It is left to the writer of Proverbs to suggest moral grounds for abstaining, as well as moderation."⁵

Erman saw the main change in 1924:  the translator "ersetzte ... »den Gott« durch seinen Jahwe," replaced "the god" with his Yahweh.¹⁰  The point needs one limit.  Amenemope's "the god" often reads like a single high god, and Kidner concedes that the Egyptian "sometimes rises to heights that are worthier of an Israelite and a Christian than of a polytheist."²  So the contrast is not between a moral Hebrew and an immoral Egyptian.  It is between a pious Egyptian who also prayed to the sun and honoured Fate and Fortune, and a Hebrew editor who allowed no god but the LORD.

Part 8: Inspiration, the headings and the date

What the book says about its own sources

The headings of Proverbs name more than one author.

The quotations are from the NET.⁸  Fox reads the Hebrew of 24:23, lahakamim, as "a typical use of the lamed of authorship," that is, "These too are of the wise."  He notes that the Greek translators and some medieval commentators obscured the phrase "In order to maintain Solomonic authorship."⁴  The Hebrew text itself made no such effort.  It says plainly that this section comes from "the wise," and the wise were not all Israelites.

Does borrowing conflict with inspiration?

The skeptic's argument runs:  if a book is God's word, it cannot be copied from a pagan book.  That premise is not the Bible's own.  Second Timothy 3:16 says "Every scripture is inspired by God," which is a claim about the result, not about where the writer found his material.⁸  The Bible quotes non-Israelite sources and says so.  Paul quotes Greek poets:  "as even some of your own poets have said, 'For we too are his offspring'" (Acts 17:28), and a Cretan prophet (Titus 1:12).  Jude 14 cites a prophecy of Enoch from outside the Hebrew canon.  Acts 7:22 says Moses "was trained in all the wisdom of the Egyptians."⁸

The conservative scholars who accept the borrowing make this point from inside the tradition.  Kitchen writes that "'authorship' of wisdom-compositions of this kind does not necessarily imply personal invention of everything that is presented," and that ancient authors "were as often collectors and reformulators of treasured lore as actual inventors."²⁰  Waltke, who takes Amenemope as the "structural model" of the section, still writes of the sage's teachings as "inspired, and thus trustworthy."²⁸  Kidner's image is the plunder of Egypt at the Exodus:  "Egyptian jewels ... have been re-set to their advantage by Israelite workmen and put to finer use."²

A borrowed source would conflict with inspiration in two cases:  if the book claimed these sayings were revealed to Solomon alone, or if it took over the Egyptian theology.  It does neither.  It names "the wise" as the source, and Part 7 shows it removing the Egyptian gods.

What the parallel says about the date

The parallel gives a floor and no ceiling.  The Hebrew section cannot be older than Amenemope, so it is later than about 1100-1000 BC on most Egyptologists' dates.  But Amenemope was copied in Egyptian schools for some five centuries after that, so a Hebrew writer could have met it at any time from Solomon to the exile.

1. Solomon, about 950 BC. Ruffle: "I really do not see why this work could not be undertaken if not by Solomon at least by a scribe working directly under his supervision."⁵ Kitchen reads the book's headings as "the explicit literary history of the existing book."²⁰ Waltke ascribes the thirty sayings to Solomon.²⁸ The heading of 1:1 covers the whole book, and on this view Solomon gathered "the words of the wise" into it.
2. Hezekiah's century, about 700 BC. Fox holds that "Only the period from the end of the eighth century to the exile provided the preconditions for literary transmission."⁴
3. Not Persian. Even Fox rejects Erman's Persian-period route, because it would place the section "too late to fit the character of the language and the indicators of a monarchical setting," and because "two maxims (22:29 and 24:21-22) speak to persons in the royal court."⁴ Simpson placed the section "in the latter part of the Monarchy."¹²

So the scholars read for this article agree that the section belongs to the monarchy, and they disagree within it.  Amenemope cannot settle that disagreement.  A reader who wants Proverbs 22:17-24:22 to be Solomonic, or to be late, must argue from other evidence.

Objections and responses

"If Proverbs copied an Egyptian book, it is not inspired."

Inspiration has never meant that a biblical writer invented every sentence.  The section says what it is:  "the words of the wise," followed by "These sayings also are from the wise."⁸  The same book includes the sayings of Agur and of King Lemuel, and First Kings measures Solomon's wisdom against "all the sages of Egypt."⁸  Paul quotes pagan poets in Acts 17:28.  What a believer claims is that God guided the selection and the use.  In Proverbs 22-23 the use is visible:  the sayings are re-founded on trust in the LORD, who "will plead their case."  Denying the Egyptian source in order to protect inspiration protects a doctrine the Bible does not teach.  Conservative commentators from Kidner to Waltke accept the source and the inspiration together (Part 8).

"Solomon came first, so the Egyptian must have copied Proverbs."

This needs Amenemope to be younger than Solomon, and it is not.  The earliest copies are school exercises from about the 21st or 22nd Dynasty, and the scholars read here who knew those copies place the composition between roughly 1300 and 950 BC (Part 2).  The theories that made the Egyptian the borrower, Kevin's and Drioton's, put Amenemope among the Jewish colonies of the exile or the Persian period, and the manuscripts rule that out.  The Egyptian text also contains puns that work only in Egyptian.⁷  A defender who uses this argument is repeating a position its own best advocates abandoned.

"These are coincidences. Wise men everywhere say the same things."

For many of the pairs this is true, and Ruffle showed it with Aztec precepts.⁵  Table manners, care for the poor and warnings against anger are universal, and Part 3 grades those pairs as partial or theme only.  But coincidence does not explain the cluster:  21 of 24 verses in one block, 1 of 46 in the next.  It does not explain a Hebrew book that refers to "thirty" sayings beside the one Egyptian instruction known to number its thirty chapters.  And it does not explain riches that "make wings" and fly "into the sky" in both books, an image built the same way in each.

"The "thirty" is an emendation made to fit Amenemope."

The reading "thirty" was proposed by Felix Perles in 1905-1906, nearly twenty years before Amenemope was known.⁴  The written word is the only one of fourteen uses of shilshom in the Hebrew Bible that stands without "yesterday," the qere means "officers," and the Greek translators already saw "three" in it (Part 4).  "Thirty" changes vowels only.  Ruffle's own alternative, "previously," adds a word.

"If the Bible borrowed here, it borrowed everything, and the whole book is human."

The borrowing is measurable and small.  It is concentrated in 22:17-23:11, and Simpson finds one Egyptian parallel in the 46 verses that follow.  Kidner counts about seven more scattered through the rest of Proverbs.²  Fox traces four "better than" sayings to Amenemope's template.⁴  Against a book of 915 verses, that is a small, well-defined use of one foreign source.  It is evidence that Israel's sages read foreign books, which the Bible says itself.  It is not evidence that Israel's faith came from them, and Part 7 shows the opposite.

"The borrowing proves the section is late, not Solomonic."

It proves neither.  Amenemope was older than Solomon and was still being copied in the sixth century BC, so it fits any date in the monarchy.  Fox's argument against Solomon rests on his view that the books of Kings are "semi-legendary" for that period, not on Amenemope.⁴  A reader who trusts the account in Kings has a plausible Solomonic route;  a reader who does not has a plausible Hezekian one.  The parallel cannot decide between them.

"A Judean could not read Egyptian, so Ruffle's model fails."

Fox makes the point about a Judean.  Ruffle's model is an Egyptian scribe at the Judean court, which is what Mettinger's reading of the title of David's and Solomon's secretary would suggest.²⁹  Fox's alternative, an Aramaic translation, is also unattested.  Both models need an intermediary no one has found.  Neither affects the direction of borrowing.

"This only shows that Israel's ethics came from Egypt."

Partly true, and a defender should say so.  Israel shared a moral vocabulary with its neighbours, and Proverbs used an Egyptian book to teach it.  But Israel's own law already forbade moving a boundary stone (Deut 19:14, 27:17), so the borrowed saying fit a moral rule Israel held on other grounds.⁸  And what Proverbs changed was not the ethics but the reason for them:  "So that your confidence may be in the Lord."

Issues and unanswered questions

1. Laisney's edition was not read. Vincent Laisney's L'Enseignement d'Aménémopé (2007), the current critical edition with the fullest list of manuscripts, has no open copy, and the reviews of it were not reached. The manuscript table in Part 2 rests on Ruffle, Williams, Grumach, Griffith, Lange, Budge and Fox.
2. The early dates rest on reports. Černý's date for the Cairo ostracon is known only from his word as Ruffle and Williams report it; he never published it, and Ruffle says the sherd is now untraceable. Peterson's 1966 article on the Stockholm papyrus and Posener's 1966 article on the writing boards were not read, and Fox reports Peterson's date more widely (21st-26th Dynasty) than Grumach does (21st-22nd). If the earliest copies are later than reported, the composition date rests on Washington's names, loanwords and chapter headings, which still point to the 20th Dynasty.
3. The British Museum's own record for EA 10474 sits behind a bot check and was not read. The four published dates for the papyrus span the 20th to the 27th Dynasty.
4. Ruffle and Griffith contradict each other on the scribe's name. Ruffle says the name Senu "is not otherwise found after the XXth Dynasty"; Griffith says "the only other instance known of Senu is Ptolemaic."¹⁴ Neither is used here to date the papyrus.
5. Waltke was read only through Google Books snippets. His view that Solomon composed the thirty sayings on Amenemope's model is clear; his fuller argument in Proverbs 1-15 was not reached. The word "appropriated," sometimes attributed to him, was not found in the snippets and is not attributed to him here.
6. Not read: Gressmann 1924, Humbert 1929, Whybray 1994, Emerton 2001, Fox 2014, Shupak 2005, Römheld 1989, Overland 1996, Bryce 1979, Goldwasser 1991 on hieratic in Judah, and Lambdin 1953 on Egyptian loanwords. Gressmann's and Whybray's positions are stated through Ruffle and Fox. Simpson 1926 was read in a badly damaged scan and is quoted only in a few words.
7. The counts are scholars' lists. The cluster and order figures in Part 3 are computed from Simpson's table. A table built by someone looking for parallels is not a random sample. A fair test would have a scholar who has not seen the Egyptian grade every verse of Proverbs against Amenemope blind. No one has done it.
8. Kitchen's view of the Amenemope link is not stated in the works read. He deferred to Ruffle in 1977, and his 2003 book mentions Amenemope only in passing. This article does not count him in the common-source camp.

Verdict

Proverbs 22:17-23:11 depends on the Instruction of Amenemope, and the dependence runs from Egypt to Israel.  The parallels pass all four tests:  close wording in about seven places, an unusual shared detail in the number thirty and the winged riches, an Egyptian text older than any date for the Hebrew, and a plausible route on either a Solomonic or a Hezekian date.  The parallels cluster in 24 verses and stop.  The theories that the Egyptian copied the Hebrew rested on a late date for Amenemope that the manuscripts disproved, and the theory of a common source has no manuscript.

The use was free and selective.  The order is mostly rearranged.  About ten Egyptian chapters were drawn on, cut short, set among sayings from other sources, and counted into thirty Hebrew sayings.  Ruffle's dissent is right that the copying was looser than early scholars claimed, and it still leaves the material coming from Amenemope.

The theology was not borrowed.  The Hebrew writer removed Fate, Fortune, the underworld and the prayer to the sun, and added the LORD as the reason for the teaching and as the advocate of the poor.  The section calls itself "the words of the wise."  A reader who holds that God inspired Proverbs has nothing here to explain away;  a reader who holds that Proverbs is merely borrowed Egyptian wisdom has not read what the Hebrew writer did with it.

The parallel does not date the section beyond the monarchy.  It is consistent with Solomon and consistent with Hezekiah.

Sources