GAGARIN, MICHAEL, The Organization of the Gortyn Law Code , Greek, Roman and 
Byzantine Studies, 23:2 (1982:Summer) p.129 

The Organization of the Gortyn Law Code 

Michael Gagarin 

I T IS GENERALLY MAINTAINED that the great inscription of laws at 
Gortyn1 is not a code in the strict sense of a comprehensive and 
systematic body of laws but is simply a collection of laws on vari­

ous subjects rather haphazardly conjoined-a satura legum, as the 
editors of JJG (p.441) call it. Nevertheless, the collection is arguably 
a code in the more general sense of an authoritative publication of 
laws;2 as I hope to show, moreover, we find in this Great Code (as it 
is often called) a certain organization of provisions into main sections 
devoted to various subjects and subsections within these. I agree with 
most scholars that the legislation of the Code is for the most part 
earlier than this particular publication,3 and I shall suggest that the 
historical process by which laws were originally inscribed separately 
and then reinscribed on the Great Code accounts for and is revealed 
in this organization. 

My case will be based on the punctuation of the inscription by 
means of asyndeton and gaps of one or two letter spaces. This 
method of punctuation is a relatively late (fifth-century) development 
in Gortynian inscriptions, and nowhere else is it used so extensively. 
In the earliest period the primary mark of punctuation is a vertical 
line dividing words and groups of words~ occasionally a special Iigo 
(t><l) is used to mark a new provision.4 These vertical dividers appar­
ently became less frequent5 and are no longer found at Gortyn after 

I The standard edition is M. GUARDUCCI, I.Cret. IV 72. Other editions here­
after referred to by the editor's name alone are by D. COMPARETTI in MonAnr 3 
(1893) 93-242; R. Dareste, B. Haussoullier, Th. Reinach, Inscriptionsjuridiques grecques 
I 352-493 (hereafter' IJG'); F. BLASS, SGDI III 4991; C. D. BUCK, The Greek 
Dialects (Chicago 1955) no. 117; R. F. WILLETTS, The Law Code of Gortyn (Berlin 
1967). 

2 See M. Lemosse, "Les lois de Gortyne et la notion de codification," RIDA IlIA 
(1957) 131-37. 

3 E.g. Willetts 8-9. 
4 For the dating of all Cretan inscriptions I rely on L. H. Jeffery, The Local Scripts 0/ 

Archaic Greece (Oxford 1961) 309-16. The ligo is found on inscriptions from Dreros 
(BCH 61 [1937] 333-48 [Meiggs/Lewis 2]) and Lyttos (J.Cret. I xviii 5). 

5 One inscription (J.Cret. IV 13) has two lines of writing, the second presumably 
later than the first. The first line has eleven vertical dividers and the second has only 
three. 

129 



GAGARIN, MICHAEL, The Organization of the Gortyn Law Code , Greek, Roman and 
Byzantine Studies, 23:2 (1982:Summer) p.129 

130 THE ORGANIZATION OF THE GORTYN LAW CODE 

the earliest group of laws (J.Cret. I 1-40).6 Asyndeton at the begin­
ning of a new provision is found twice in an early inscription from 
Gortyn.7 After ca 500 (I.Cret. IV 41-140) asyndeton (usually pre­
ceded by a gap) becomes common, as do gaps between provisions 
without asyndeton.8 It may be no accident that these means of punc­
tuation become common just as the Gortynians are beginning to 
collect groups of laws together. I shall argue, in fact, that this punctu­
ation preserves a record of the historical development of these collec­
tions, as can most easily be seen on the Great Code. 

I. Asyndeton 

All editors of the Code recognize that new sections are marked by 
asyndeton, normally preceded by a gap of one or (rarely) two spaces.9 

But in their discussions all editors to some extent ignore the guidance 
of asyndeton in dividing the Code into sections. Willetts (4), for 
example, speaks of asyndeton as marking either "a wholly new topic" 
or "a partially novel topic," without specifying the nature of this 
distinction. I shall argue that every asyndeton marks the beginning of 
a new section and indicates that the provision or group of provisions 
it contains was originally enacted or assembled as a unit. In many 
cases these units were inscribed elsewhere and later reinscribed on 
the surviving inscription.lo 

Following the guide of asyndeton we may divide the Code into the 
following sections: 

6 An intermediate form of punctuation by means of a colon (:) is found only in one 
small fragment, I.Cret. IV 62, which together with 63 and 64 is dated ca 525-500. 
There are no gaps in these inscriptions. 

7 I.Cret. IV 14.g-p. In both cases the new provision may be unrelated to what 
precedes. 

8 I.Cret. II xii 15.6, with gaps before and after a single word, is unparalleled in early 
inscriptions. 

9 Only three instances of asyndeton introducing a new section are not preceded by a 
gap 0.37, 5.1, 8.30) and these may be the result of scribal error. Three other cases 
should be mentioned: the asyndeton in 6.1 is in fact preceded by a small gap at the end 
of 5.54, not noticed by most editors (compare the gap at the end of 5.34); in 6.55 I 
take TO EAEV(Jipo with the preceding sentence (so Blass); in 9.1 there is a break in the 
stone but it seems certain that there would have been a gap of at least one and perhaps 
two letters. 

10 For convenience I shall not discuss the possibility that one or more intermediate 
stages of reinscription may have occurred. Such an occurrence would not significantly 
affect my views. The oral transmission of laws at an earlier period, suggested by Wil­
letts (Aristocratic Society in Ancient Crete [London 1955] 5-6), is extremely doubtful; 
even the earliest Cretan laws show no trace of oral composition. 



GAGARIN, MICHAEL, The Organization of the Gortyn Law Code , Greek, Roman and 
Byzantine Studies, 23:2 (1982:Summer) p.129 

MICHAEL GAGARIN 

(1) 1.2-2.2: Seizure of Persons (57 lines) 
(2) 2.2-10: Rape (8) 
(3) 2.11-16: Forcible Intercourse with a Slave (5) 
(4) 2.16-20: Attempted Seduction (4) 
(5) 2.20-45: Adultery (25) 
(6) 2.45-3.16: Divorce (26) 
(7) 3.17-37: Separation of Spouses (21) 
(8) 3.37-40: Special Payments to a Spouse (3) 
(9) 3.40-44: Separation of Slaves (4) 

(10) 3.44-4.8: Children of Divorced Women (19) 
(11) 4.8-17: Exposure of Children (9) 
(12) 4.18-23: Unwed Slave Mothers (5) 
(13) 4.23-5.1: Distribution of Property among Children (32) 
(14) 5.1-9: Non-retroactivity of Law on Gifts to Women (8) 
(15) 5.9-54: Inheritance and Division of the Estate (45) 
(16) 6.1-2: Gifts to a Daughter (1) 
(17) 6.2-46: Sale and Mortgage of Property (44) 
(18) 6.46-56: Ransom of Prisoners (0) 
(19) 6.56(?)-7.10: Marriage of Slave Men and Free Women (10) 
(20) 7.l0-15: Liability of a Master for his Slave (5) 
(21) 7.15-8.30: Marriage or Remarriage of the Heiress (70) 
(22) 8.30-9.1: Further Provisions concerning Heiresses (26) 
(23) 9.l-24: Sale or Mortgage of Heiresses' Property (23) 
(24) 9.24-40: Liability of Heirs (16) 
(25) 9.40-43: The Son as Surety (3) 
(26) 9.43-10.?: Business Contracts (11 +) 
(27) 10. ?-25: Gifts of Males to Females (10+) 
(28) 10.25-32: Restrictions on the Sale of Slaves (7) 
(29) 10.33-11.23: Adoption (43) 
(30) 11.24-25: Amendment to Section 1 (1) 
(31) 11.26-31: The Duty of Judges (5) 
(32) 11.31-45: Amendment to Section 24 (14) 
(33) 11.46-55: Amendment to Section 6 (9) 
(34) 12.1-5: Amendment to Section 27 (4) 
(35) 12.6-19: Amendment to Section 22 (13) 

131 

Several of these divisions, indicated by asyndeton, are not recognized 
by one or more editors of the Cadell and require a brief discussion. 

The second column contains provisions on rape (2.2-10), forcible 
intercourse with a household slave woman (2.11-16), attempted se­
duction (2.16-20), and adultery (2.20-45). These sections are obvi­
ously related and are usually grouped together into one large section, 
but asyndeton indicates (in my view) that each was enacted sepa-

11 ] refer primarily to the divisions of Willetts (summarized on p.34) and Guarducci 
(summarized on p.149). 



GAGARIN, MICHAEL, The Organization of the Gortyn Law Code , Greek, Roman and 
Byzantine Studies, 23:2 (1982:Summer) p.129 

132 THE ORGANIZATION OF THE GORTYN LAW CODE 

rately. Separate enactment is further indicated by the lack of integra­
tion of the provisions. For instance, the law against 'forcible inter­
course' with a household slave woman is not a part of the more 
general law against rape; these provisions are placed next to each 
other but they use different terms for what was quite likely the same 
offense12 and they set quite different conditions for determining the 
fine to be paid. Similarly, the law against attempted seduction, which 
follows, describes a set of circumstances for the offense different 
from those in the preceding and following sections, indicating that it 
too was originally a separate enactment.I3 Thus it is likely that all four 
of these sections were separate enactments and are so marked by 
initial asyndeton in each case. 

The third column contains three instances where the beginning of 
a new section, marked by asyndeton, is recognized by Willetts but 
not by Guarducci: 3.16, 3.37 (where there is no gap), and 3.40. 
Though here too the provisions are related (they all treat the affairs 
of a husband and wife), they regulate quite different matters: divorce 
(2.45-3.16), the death of one spouse (3.17-37), komistra (probably 
payments from one spouse to another 3.37-40),14 and the separation 
of slave couples (3.40-44). As in column two, the provisions on 
death and divorce, which are obviously related, are presented as 
separate matters with no attempt at integrating them into a coherent 
whole. And the provision concerning komistra, though it too concerns 
the financial affairs of a husband and wife, apparently does not refer 
to either divorce or death. Thus the initial asyndeton here too indi­
cates that this provision was conceived as a separate section. 

These three sections are followed by a brief one (3.40-44) concern­
ing the separation of slave couples by divorce or death. In contrast to 
the preceding sections concerning free spouses, the separation of 
slaves is treated in a single integrated section. Since it is likely that 
disputes concerning property in the case of free persons came to the 
attention of lawmakers earlier than those concerning the property of 
slaves, it is plausible to see 3.40-44 as a separate later enactment. 

12 No one has plausibly suggested any difference between the acts of 'rape' (KapTH 
Ot1TiL, 2.3) and 'forcible intercourse' (KapTH oa~amTo, 2.11-12). 

13 Willetts' long note on this difficult provision (2.16-20) is helpful, but his claim 
that it forms "a logical bridge between the regulations about rape and those about 
adultery" is unconvincing. Neither the sequence of offenses (rape, attempted seduc­
tion, adultery) nor the fines (100, 10, 100 staters) reveals any logical sequence. 

14 The precise nature of komisfra is not clear, but the purpose of restricting the 
amount of these payments is probably to protect the interests of the heirs of either 
spouse. Willetts' view that they are payments for 'porterage' is doubtful, since there 
seems to be no reason why a wife should pay for her husband's porterage. 



GAGARIN, MICHAEL, The Organization of the Gortyn Law Code , Greek, Roman and 
Byzantine Studies, 23:2 (1982:Summer) p.129 

MICHAEL GAGARIN 133 

Thus 2.45-3.44 consists of four separate sections, each of which is 
marked by initial asyndeton. 

There follow a number of provisions concerning children of single 
(unwed or divorced) women (3.44-4.23). Though there is asyndeton 
in two places in these provisions (4.8, 4.l8), Willetts treats them all 
as one section (Guarducci divides it at 4.l8). Here too, however, I 
think the asyndeton may be significant. The first section (3.44-4.8) 
states the procedure by which a divorced woman who bears a child is 
to give her former husband (or his master in the case of a slave) the 
choice of keeping the child; the second section (4.8-17) states the 
penalties for the unlawful exposure of a child by a divorced woman. 
The third section (4.l8-23) gives the rules for ownership of the child 
of an unmarried slave woman, and is clearly separate from the first 
two. Though the second of these sections sets the penalties for viola­
tions of the rules set forth in the first section,15 it may originally have 
been written as a separate section, since in addition to the asyndeton 
the section begins unusually with two words indicating the subject of 
the following provisions (yvva K€peVovan). When such words pre­
cede the conditional term, this normally indicates the beginning of a 
new section.16 Thus even though 4.8-17 was written in full awareness 
of the preceding section, it may have been enacted separately, provid­
ing sanctions for the regulations (without sanctions) which had al­
ready been enacted.17 

The next problematic caselS is the asyndeton (without a gap) in 
5.l, where, after regulations specifying the distribution of property to 
sons and daughters, a provision is made limiting the retroactivity of 
provisions concerning gifts to women (5.l-9): a woman who had 
received no property as a gift, a dowry, or an inheritance from the 
time when Kyllos was kosmos was to obtain her due portion, but for 
women before this time there was no legal recourse (i.e., women 
who had not received a proper share before this time could not now 
sue to obtain it) .19 To the extent that this provision limits the retro-

15 KaTel Tel t'YpaJ.t~va in 4.10-11 presumably refers to the regulations in 3.44-4.8. 
16 Words indicating the subject of the provision to follow also begin sections at 2.11, 

3.37, 5.1, 6.1, and 9.30. 
17 For the distinction between a regulation and a sanction see my Drakon and Earlv 

Athenian Homicide Law (New Haven 1981) 81-82. . 
18 At 4.27 Willetts and Guarducci punctuate with a full stop followed by asyndeton. 

Other editors mark only a half stop, which is clearly correct (see in/ra n.27). 
19 The interpretation of this section is disputed~ I plan to consider it at length else­

where. Willetts' interpretation seems motivated primarily by his general theory that 
women were more restricted by the provisions of the Great Code than they had been 
earlier. This theory seems to me dubious, though I cannot undertake to examine it 
here. 



GAGARIN, MICHAEL, The Organization of the Gortyn Law Code , Greek, Roman and 
Byzantine Studies, 23:2 (1982:Summer) p.129 

134 THE ORGANIZATION OF THE GORTYN LAW CODE 

activity of the immediately preceding provisions, it might perhaps 
have been joined with them, but in fact it has a considerably wider 
scope, since it applies to gifts and dowries in addition to inheritance. 
Thus we should again follow the guidance of the asyndeton, which 
indicates that 5.1-9 was a separately enacted section20 placed next to 
one of the set of regulations to which it refers.21 

The last two problematic cases of asyndeton (8.30, 9.1) occur in 
what is usually considered one long section on the treatment of 
heiresses22 (7.15-9.24). The initial section (7.15-8.30) contains a se­
ries of prescriptions for the marriage of an heiress, whether unmar­
ried or already married. After asyndeton in 8.3023 we find an as­
sortment of different regulations pertaining to heiresses (8.30-9.1). 
Though these regulations must have been written with reference to 
those in 7.15-8.30, it is possible to see them as a later enactment. 
That there was no attempt to integrate the later rules into the earlier 
section is clear from the presence in the later section of the definition 
of the heiress (8.40-42), which would presumably be put at or near 
the beginning of an integrated section. Thus it is likely that here too 
the asyndeton indicates a separate enactment. 

The regulations which follow the next asyndeton in 9.1 concern the 
management of the heiress' property, giving her a certain amount of 
control over it (9.1-24). These provisions have no direct connection 
with any of the preceding provisions, and they are also related to the 
provisions concerning the sale and mortgage of property in 6.2-46. 
They form a complete and coherent section in themselves, and there 
is every reason to see them as a separate enactment, as indicated by 
the asyndeton. 

From this examination we may conclude that in all cases asyndeton 
provides a reasonable guide for the division of the Code into sepa-

20 This section (5.1-9) was not necessarily enacted at the time of the inscription of 
the Code, since the provision for retroactivity would still have effect a number of years 
after its original enactment. 

21 Note also that this section begins with a word indicating its subject matter (see 
supra n.16). 

22 I use the traditional English translation 'heiress' even though it does not corre­
spond precisely to the Greek 'Ttn'Tp{)LOKO<;, literally 'the holder of the paternal estate'. 
At Gortyn a woman was an heiress only when she had no living father or brother (see 
8.40-42). 

23 There is disagreement about the reading at 8.30. Comparetti cites Halbherr's state­
ment that the N in ANEP seems to be a M m or a double N. Guarducci (followed 
by Willetts) says there is a fault in the stone between Nand E. Examining the stone I 
found no fault in it but rather an apparent reduplication of the initial A (AANEP). The 
crossbars on both A's are faint, and thus AN could easily be read as a M. Emending 
AANEP to KANEP (= Kat. Cxlllp) would remove the asyndeton. 



GAGARIN, MICHAEL, The Organization of the Gortyn Law Code , Greek, Roman and 
Byzantine Studies, 23:2 (1982:Summer) p.129 

MICHAEL GAGARIN 135 

rately enacted sections. Though the precise reasons for the separate 
enactments are unknown, they presumably reflect historical condi­
tions. Provisions or groups of provisions were enacted as a need was 
perceived. In several cases I have suggested reasons for the later 
enactment of a section, and other sections could be similarly ex­
plained. The significant point. however, is that although separately 
enacted sections are sometimes placed next to related sections, their 
provisions are not integrated. In other words, the Gortynians do not 
appear to have undertaken any substantial revision of their laws when 
they reinscribed them on the Great Code. 

The structure revealed by asyndeton on the one hand confirms the 
traditional view that the Code is not a systematic and comprehensive 
set of laws~ on the other hand, however, it reveals a greater organiza­
tion of provisions than is sometimes acknowledged. After the regula­
tions about seizure of persons in the first column, the next eight 
columns (2.2-9.40) regulate in succession the general areas of of­
fenses against women, relations between spouses, children, familial 
property, and inheritance. Only three short provisions in 6.46-7.15 
interrupt the general continuity of this part of the Code. Four more 
brief provisions (9.40-10.32) in no particular order form the end of 
the main body of the Code. They are followed by supplementary 
provisions (10.33-12.19), many of which are amendments to the 
main text. Thus about two-thirds of the main body of the Code 
consists of regulations about the family and its property, and some 
attempt was clearly made to group together related sections. 

A similar organization is observable on other legal inscriptions at 
Gortyn.24 Though in many cases the inscriptions are too fragmentary 
to allow for certainty, asyndeton (usually preceded by a gap)25 seems 
always to mark a new section, sometimes apparently related to what 
precedes (41.11.16, 41.111.7, 77.B.4, 77.B.I0) and sometimes appar­
ently not (14.g-p, 42.B.9, 46.B.6, 75.C.3, 76.B.7).26 As far as we can 

24 I include in this survey all the inscriptions in J. erer. IV 1-140. 
25 In two cases in one early inscription (14.g-p.1-2) asyndeton is not preceded by a 

gap. This inscription contains numerous inconsistencies in punctuation (a dividing 
mark at the second asyndeton but not the first), syntax (variation between nominative 
and accusative in two places), and use of elision (in line 1), and I would hesitate to 
draw any conclusions from it. 

26 I do not include in this list three cases where Guarducci's text has an asyndeton. 
One inscription (47) begins ... KUTaKEiJ-LElJo<;. ui K' a8tK-y)t:TEt 8011.0<; KTA. It is more 
likely, however, that KUTaKEiJ-LE IJO<; is part of the conditional clause (modifying 8011.0<;). 
The law clearly concerns a slave "given as a pledge," and since the top of the inscrip­
tion is apparently preserved, KUTaKEi/-LEIJO<; is likely to be the first word in the law 
(preceded by a gap or perhaps by 0). The reason for beginning the law with this parti-



GAGARIN, MICHAEL, The Organization of the Gortyn Law Code , Greek, Roman and 
Byzantine Studies, 23:2 (1982:Summer) p.129 

136 THE ORGANIZATION OF THE GORTYN LAW CODE 

tell, the use of asyndeton on these early inscriptions is consistent 
with what we have observed on the Great Code.27 

The early inscriptions from Gortyn also give us a glimpse of the 
development of the manner of inscribing laws. Several contain more 
than one line of retrograde script and these have been plausibly ex­
plained by Jeffery as separate inscriptions each beginning from the 
right rather than a continuous retrograde text.28 I. Cret. IV 1, for 
example, has five lines of writing, each of which appears to be in a 
separate hand, suggesting that each line probably contained a sepa­
rately inscribed (and separately enacted) law. Presumably each new 
law began with asyndeton. When the Gortynians later made the 
important change from extremely long lines of writing to the more 
convenient form of columns,29 separately enacted laws could be in­
scribed in a separate column or, if in the same column, separated by 
a space. Several inscriptions show an empty space below the written 
text, and one stone (I. Cret. IV 43) is particularly interesting for its 
organization of separate provisions. The inscription is in two columns 
of different widths, each containing two sections separated from each 
other by an empty space. Each of the four sections concerns a differ­
ent subject: the rental of a threshing floor (Aa) , the treatment of a 
slave (Ab) , the use of a public orchard (Ba), and the use of water 

ciple is to indicate immediately its subject matter (see supra n.16). In the second case 
(42.8.3) asyndeton occurs at the beginning of a rather long space (19 letters) in which 
the original text was erased and a new text inscribed. Since this provision seems to 
follow the preceding one closely, I suspect it may originally have contained a connect­
ing particle, which had to be omitted when a corrected text was inscribed in the limited 
space of the erasure. In the third case (7S.3-4) Guarducci's restorations seem too long 
for the gap and may well be wrong. Other restorations could easily include a connecting 
particle. 

27 In addition there are perhaps four examples of a special kind of asyndeton, where a 
general introductory statement or proclamation is followed by a more specific provision 
introduced by asyndeton. For example, 43.Ba begins with the declaration, "the city has 
given the orchard (or vineyard) in Keskora and in Pala for planting" 0-3), and there 
follow, with asyndeton (but no gap), two provisions (3-9) regulating the use of these 
fields. Here and in three other examples (41.1.5, 7S.1, SO.1), where an initial statement 
is followed by more specific regulations, one might reasonably punctuate with a half 
stop (equivalent to our colon) rather than a full stop. A similar situation obtains in 4.27 
of the Great Code, where an initial statement of the parents' authority (4.23-27) is 
followed by regulations for the division of their property. These regulations begin with 
asyndeton (without a gap), but some editors UJG, Comparetti, Blass) do not punctuate 
with a full stop. 

28 Jeffery (supra n.4) 312. 
29 We need not explain this change as influenced by the columns of writing on papy­

rus, as Guarducci (p.S7) does. The proliferation of laws would in itself soon force the 
Gortynians to realize the greater convenience of columns. 



GAGARIN, MICHAEL, The Organization of the Gortyn Law Code , Greek, Roman and 
Byzantine Studies, 23:2 (1982:Summer) p.129 

MICHAEL GAGARIN 137 

from a public stream (Bb). The first two of these (Aa, Ab) have 
references to other laws which must have been inscribed elsewhere, 
perhaps on a nearby stone. The other two (Ba, Bb) seem to be self­
contained provisions, and the fact that each begins with the common 
invocation BLOt confirms that they were enacted separately. Clearly 
these are four separately enacted laws, brought together primarily 
because there was room on the stone, though the two laws in column 
B may have been thought of as related in content. Had the occasion 
arisen, each law might have been reinscribed elsewhere with other 
provisions. 

Other laws were gathered together and reinscribed in groups, as on 
the Great Code. Each section was reinscribed as it originally began, 
without any connecting particle and separated from the preceding 
section by a small gap. A number of Gortynian inscriptions show 
traces of earlier writing, and though we cannot be certain that the 
earlier inscriptions were all laws, it is likely that they were.30 These 
earlier traces indicate that the stones were erased, presumably at the 
time when the laws on them were reinscribed elsewhere. On some 
stones (e.g. I. eret. IV 47) it is immediately evident that the writing 
surface is indented, presumably because an earlier inscription has 
been erased. 

One other method of inscribing new laws is shown in the provi­
sions added at the end of the Great Code (10.33-12.19). With one 
exception each of these begins on a new line, leaving a large gap at 
the end of the preceding line but no space between lines.31 The first 
of these, in the same hand as the rest of the Code,32 concerns adop­
tion and was probably inscribed soon after the main body of the 
Code. The remaining provisions were added later. Except for this 
section and one on the duty of judges (11.26-31), which refers gen­
erally to the whole Code, each supplementary provision is clearly an 
amendment to a particular section of the Code. Presumably these 
amendments were inscribed here as they were enacted. 

30 The letters discernable in I.Cret. IV 48, for example, probably form part of a law or 
decree; see Guarducci ad loc. 

31 Note that several of these added sections begin from the left 01.26, 11.46, 12.6). 
This is contrary to the normal Cretan practice that boustrophedon inscriptions begin 
from the right, but the fact that I. Cre!. IV 43.Bb also begins from the left indicates that 
in the fifth century this rule did not apply consistently to new laws inscribed after a 
previously inscribed law. 

32 According to Guarducci 10.33-11.23 are in the same hand as the main body of the 
Code but 11.24-12.19 are in a different hand. I treat these all as later additions because 
of the large gap in 10.32. 



GAGARIN, MICHAEL, The Organization of the Gortyn Law Code , Greek, Roman and 
Byzantine Studies, 23:2 (1982:Summer) p.129 

138 THE ORGANIZATION OF THE GORTYN LAW CODE 

II. Gaps Within Sections 

In addition to the division of the Great Code into sections by means 
of asyndeton, there are a large number of gaps (about fifty) not fol­
lowed by asyndeton. Most of these mark subsections within the main 
sections and enable us, I shall argue, to see the internal organization 
of a section, which in many cases may have been determined by 
historical circumstances. Willetts (4) maintains that "a vacat occurs 
elsewhere without asyndeton, presumably where the engraver pre­
ferred a smoother surface." This explanation is impossible, first be­
cause it is highly unlikely that so many rough spots would coincide with 
the beginning of a sentence when fewer than ten gaps (caused by a 
rough surface) occur in the middle of a sentence,33 and second because 
an inspection of the stone shows that in almost all the gaps occurring at 
the beginning of a sentence the surface is smooth. It seems likely, 
then, that the gaps within sections have some significance.34 

In the first section of the Code concerning the illegal seizure of 
another person, the first ten lines (1.2-12) state the basic rule (no 
one is to be seized before a trial), the penalty for its violation (a set 
of fines), and the method for determining the fine (the judge is to 
decide on oath). These are the most general provisions concerning 
the seizure of persons and this subsection is marked off from the 
following provisions by a gap.35 The next subsection (1.12-49) con­
tains more specific provisions regulating various situations that might 
arise in implementing the general law: the seizure is denied and there 
are no witnesses; witnesses disagree whether the seized man is a 
slave or free; two parties dispute the possession of a slave; the loser 
in a dispute refuses to release the seized person; after the trial a slave 
flees to a temple rather than going to the winner of the dispute.36 

33 The following gaps appear to have been caused by the rough surface of the stone: 
l.43, l.44, 6.16, 6.42, 9.28 (or this may be an erasure), and perhaps 3.26, 5.42 Gn a 
large lacuna), 7.2 (see infra n.44), and 12.9 (where sOme editors supply two letters to 
fill the gap). 

34 The widespread use of gaps in a continuous text is, as far as I know, unique to 
Gortyn. Elsewhere gaps are rare, except in a few specific places, such as (on occasion) 
to separate part or all of the preamble from the main body of a decree. 

35 Here and in a few other places a gap is filled with a painted palmula, which has no 
significance for our discussion. Traces of paint are randomly preserved both in gaps and 
in the lettering (e.g. at the top of columns five and six), and some painted signs (e.g. 
at 6.9) which could be seen when the inscription was first discovered have since disap­
peared (see Guarducci p.125). It is possible that all the gaps were originally filled with 
painted signs. 

36 I take 1.46-49 to refer to the case where a slave flees to a temple. The loser of the 
case must recover the slave and deliver him up within a year or else pay the single fines. 



GAGARIN, MICHAEL, The Organization of the Gortyn Law Code , Greek, Roman and 
Byzantine Studies, 23:2 (1982:Summer) p.129 

MICHAEL GAGARIN 139 

These provlsiOns form the second subsection, which is followed by 
three separate provisions each preceded by a gap: the first concerns 
the possibility that a seized person may die during a trial (1.49-51),37 
the second concerns disputes in which one party is a kosmos (1.51-
55), and the third allows the seizure of two special categories of 
persons 0.56-2.2). 

The gaps within the first section of the Code reveal the following 
organization: first the most general provisions concerning the seizure 
of persons, then a group of supplementary provisions, and finally 
three separate additional provisions. It is possible that all the provi­
sions in this section were enacted at the same time and that the gaps 
simply punctuate the section, but a historical origin seems more 
likely. The order of the subsections suggests a historical progression 
in that a self-contained group of basic provisions comes first and 
these provisions seem to reflect an early stage of written law, when 
the city was becoming strong enough to ask its citizens to submit 
their disputes to the legal process rather than using force or tradi­
tional means of self-help. Later, certain situations not covered by the 
basic law would arise: a man might seize someone and then claim 
that he had always possessed him, or a man might refuse to release a 
person even after a trial, and stiffer fines would then be necessary.38 
Experience with such situations would lead to the enactment of a 
group of supplementary provisions. Finally, at an even later period 
three more provisions relating to the law were separately enacted. 
The entire set of provisions was then reinscribed together on the 
Great Code. 

:17 Willetts argues that l.49-51 ("if he dies during the trial, he is to pay the single 
fine") concerns the death of a litigant; but if this were so, it would mean that the 
litigant who dies would automatically lose his case. This seems unlikely, since the 
man's heirs, who might at least temporarily inherit the disputed person, ought to be 
allowed to continue the case. Also, on Willett's interpretation, nothing is said of the 
disputed person: would he automatically go to the other litigant if one litigant died? 
Again, this seems unlikely. If the provision refers to the death of the disputed person, 
however, whether slave or free, its purpose might be to prevent someone who was 
likely to lose the case from simply killing the disputed person, thereby rendering the 
case moot and avoiding any fine. Guarducci takes the provision to apply only to the 
death of a slave, but if the law contained such a restriction, for which I can see no 
reason, it would surely have to be made clear in the text. 

38 I do not see how the provision in l.35-39 could possibly reduce the fine to one­
third after a year had passed, as some have argued. This would simply encourage some­
one not to release a person until a year had passed. If, moreover, as seems likely, the 
winner of a case received the fine directly, he could at any time accept a smaller pay­
ment and declare the matter ended. The only meaningful interpretation of the provi­
sion is that it increases the fine after a year, either tripling it or (less likely) adding 
three categories of fines together (so Willetts). 



GAGARIN, MICHAEL, The Organization of the Gortyn Law Code , Greek, Roman and 
Byzantine Studies, 23:2 (1982:Summer) p.129 

140 THE ORGANIZATION OF THE GORTYN LAW CODE 

It could be argued that this same organization into subsections 
containing first the most general provisions and then more specific 
provisions is as likely to be found in a set of laws enacted at one time 
as in a set of laws enacted over a period of time, but such a view 
cannot account for the separation of the last three provisions from 
one another. If each of these was a separate enactment, however, the 
gaps are easily explicable. A historical development thus seems more 
likely. Amendments were attached to the original provisions and their 
separate origin was indicated even after reinscription by the continued 
presence of a gap between them.39 

In most of the remaining sections of the Code a similar historical 
development can plausibly explain the division into subsections. The 
section on divorce, for example, contains provisions specifying the 
property a wife may keep and the penalty for her taking too much 
(2.45-3.12). After a gap there is added a provision specifying a fine 
for anyone who helps a woman take more property than she is 
allowed (3.12-16). A historical explanation is extremely plausible 
here. 

The divisions in the next section on the death of a husband or 
wife, though somewhat differently organized, may also have a histori­
cal explanation. First comes the case of a husband dying with children 
(3.17-24), then (after a gap) a husband without children (3.24-31), 
and finally (after another gap) a wife without children (3.31-37).40 
The organization of these subsections is different from that of the 
first column, but it is possible that here too each subsection was 
enacted as the need was perceived. The first case (the death of a 
husband with children) may have been the most likely to occasion 
disputes over property (between the children and their mother, espe­
cially if she remarries), and these provisions would thus be the ear­
liest. That these provisions were originally enacted as a response to 
disputes arising in such cases is suggested by the fact that the pre­
scribed remedy is that the matter shall be decided by a court. Later, 
disputes arose in the case of a husband dying childless and then in 
the case of a wife dying childless, and provisions were enacted to 
cover these. Provisions for the case of a wife dying with children 

39 If the separate provisions were originally inscribed without a connecting particle at 
the beginning, as is the case in the amendments to the Code, the scribe would add a 
particle at the beginning of each subsection. But it may have been more common in 
the early period to put a particle at the beginning of an amendment. The evidence of 
the earliest law from Dreros (see supra n.4) indicates that an amendment could be 
added beginning on a new line and separated by a punctuation mark, and yet contain a 
connecting particle, and this practice may at times have been followed at Gortyn. 

40 The case of a wife who dies with children is treated elsewhere (4.43-46, 6.31-36). 



GAGARIN, MICHAEL, The Organization of the Gortyn Law Code , Greek, Roman and 
Byzantine Studies, 23:2 (1982:Summer) p.129 

MICHAEL GAGARIN 141 

were not included here as they should have been if the whole section 
had been conceived as a unit, since these had meanwhile been 
enacted in connection with other sections (supra n.40). This whole 
section could have been written in a more systematic form, but the 
actual wording suggests the gradual accumulation of separately en­
acted provisions rather than a single enactment or revision. The 
absence of asyndeton after the gaps within the sections, however, 
indicates that the later enactments were considered direct amend­
ments to the earlier provisions rather than a separate section.41 

Similar patterns, similarly explicable on the basis of a historical 
development, also occur in the sections on adultery (gaps at 2.33, 
2.36),42 the children of divorced women (at 3.52), the distribution of 
property among children (at 4.48, 4.51), inheritance (at 5.28, 5.34, 
5.39, 5.44, 5.51),43 selling or mortgaging property (at 6.9, 6.12, 6.25, 
6.31), the marriage of a slave and a free woman (at 7.4),44 the mar­
riage of an heiress (at 7.29, 7.35, 7.40, 7.50, 7.52, 8.7, 8.8, 8.13, 
8.20),45 further regulations concerning the heiress (at 8.36, 8.40,46 
8.47, 8.53), selling or mortgaging the property of an heiress (at 
9.18),47 gifts (at 10.20), adoption (at 10.37, 10.39, 11.6, 11.19), and 
the affairs of a young heiress (at 12.9). 

Besides these gaps, which may be explained on historical grounds, 
two other passages require fuller discussion. In the section on adul­
tery (2.20-45) the first two gaps (at 2.27, 2.3 1) seem at first glance to 
be out of place. The passage reads as follows: 

If someone is caught in adultery with a free woman in her father's 
or brother's or husband's house, he shall pay one hundred staters; 
and if in another's house, fifty; and if with the wife of an apetairos, 

41 Obviously there is a fine line between the case of these separately enacted sub­
sections and that of the separate but related sections (indicated by initial asyndeton), 
which we examined above. To what extent this distinction was consciously noted by 
the Gortynians is difficult to tell. 

42 For the gaps at 2.27 and 2.31 see infra. 
43 For the gaps in 5.13-25 see infra. 
44 At 7.2 there is a space before at Be in which the stone is very rough. The rough­

ness may have been the cause of the empty space, in which case the gap has no signifi­
cance for our purpose, or an original letter may have been erased at this point. 

45 At 8.27 there must have been a gap where the stone is now destroyed. If the gap 
was intended to divide the sentence beginning in 8.27 from the preceding one, it would 
be an odd place for one, since the two provisions are very closely linked. But the gap 
may have been caused by the roughness at the edge of the stone at this point. 

46 In 8.40 a vertical incision just before the beginning of a new sentence must indi­
cate that the scribe forgot to leave the required gap, later noticed his mistake, and then 
used the by now obsolete dividing mark to indicate where a gap should have been left 
«(t: infra n.48). 

47 For the gap at 9.15 see infra n.50. 



GAGARIN, MICHAEL, The Organization of the Gortyn Law Code , Greek, Roman and 
Byzantine Studies, 23:2 (1982:Summer) p.129 

142 THE ORGANIZATION OF THE GORTYN LAW CODE 

ten; and if a slave with a free woman, he shall pay double. [gap] 
(2.27) And if a slave (with the wife, sister, or daughter) of a slave, 
five. And the captor is to proclaim before three witnesses to the 
relatives of the man caught in the house that they should ransom 
him within five days. [gap] (2.31) And in the case of a slave, (they 
should proclaim) to his master before two witnesses. [gap] (2.33) 
And if he is not ransomed, the captors may do with him as they 
please. 

The last of these gaps can be understood as indicating that the sub­
sequent provision was a later enactment, but the first two gaps in this 
passage seem odd. One might expect all the rules stating fines to be 
inscribed together without a gap, but separated by a gap from the 
following two rules concerning the proclamation before witnesses, 
which should not be separated from each other. The existing gaps are 
explicable, however, if we assume that all the rules stating fines 
except the last one (a slave in adultery with a slave) were originally 
followed directly by the rule concerning a proclamation before three 
witnesses. At a later time a pair of regulations was enacted concern­
ing the fine to be assessed and the proclamation to be made in the 
case of a slave's adultery. When all these provisions were rein scribed 
on the Great Code, the later two were not simply added on but were 
integrated into the earlier rules, the provision for a fine following the 
other such provisions and the rule about a proclamation following the 
other similar rule. Gaps were left before both these insertions as an 
indication of their separate origin. No gap was left before the first 
provision for a proclamation because this rule was originally attached 
directly to the group of rules preceding it. Such an explanation is, of 
course, speculative, but it seems the most plausible way of account­
ing for this apparently anomalous situation. 

The other difficult passage is found in the regulations for inheri­
tance. Here the first five provisions (5.9-28) are all separated by gaps 
(at 5.13, 5.17,48 5.22, 5.25). The first states that children, grand­
children, and great-grandchildren are to inherit. The next provisions 
state successively who should inherit if none of the preceding rela­
tives exists: after the direct descendants come the brothers and their 
descendants, the sisters and their descendants, other relatives, and 
finally household members. It is not impossible that each of these 
provisions was enacted separatelY,49 but in view of the systematic 

48 At 5.17 there is no gap, but a palmula is painted in between the two sentences. 
This palmula, which is occasionally painted in a gap (see supra n.35), must indicate that 
the scribe intended to leave a gap at this point (cf supra n.46). 

49 In this case the ~v in 5.10 would not have been part of the original law. 



GAGARIN, MICHAEL, The Organization of the Gortyn Law Code , Greek, Roman and 
Byzantine Studies, 23:2 (1982:Summer) p.129 

MICHAEL GAGARIN 143 

nature of the rules, it seems more likely that the gaps were inserted 
here in order to indicate the organization of these provisions, which 
may have been enacted all at once or over a period of time. This is 
the one passage in the Code that does not lend itself to a historical 
explanation of the gaps within sections~ it is also the most systematic 
section of the entire Code. 

Taking all this evidence together, 50 we find that the vast majority 
of gaps within sections of the Code lend themselves readily to a 
historical explanation. Certainly the gaps were intentionally placed in 
specific places and not inserted at random, since the scribe felt it 
necessary twice51 to indicate the erroneous omission of a gap. 52 And 
the lack of systematic organization (in most cases) of the provisions 
separated by gaps makes it likely that they were the product of a 
historical process rather than of a single enactment. 

Two patterns have been noted. The more common one, which is 
found in the first section, is that of a general rule or set of rules 
followed by more specific provisions, either grouped together or sin­
gly.53 The second pattern, which we noted in the section on the death 
of a spouse (3.l7-37), is that of provisions on one matter followed 
by provisions on a separate but closely related matter.54 The patterns 
are not always clearly distinguishable,55 and some longer sections 
offer a combination of the twO.56 One section (5.9-54), discussed 
above, seems to fit neither pattern, strengthening our suspicion that 
the gaps in this section may have some special significance. 57 

Despite their fragmentary nature, the evidence of the other legal 
inscriptions from Gortyn seems consistent with this interpretation of 
the gaps within sections of the Great Code. In the earliest inscriptions 

50 In addition to these gaps between sentences there are three gaps within sentences. 
One of these (9.15) occurs in regulations for mortgaging an heiress' property and 
immediately precedes a clause indicating the non-retroactivity of these regulations. 
Here a historical explanation is possible, since this clause may well have been added 
after the preceding provisions had been written. The other two gaps (at 1.16 after I/-tE// 
and 3.32 after O:7To(la//od are puzzling~ both are fairly long (about 21h letters) and occur 
at the end of clauses. I can offer no good explanation for either. 

51 At 5.17, 8.40 (see supra nn.46, 48). 
52 Of course other errors of omission (or commission) may have been left uncor-

rected. 
5:JThis pattern also occurs in 4.23-5.1, 9.1-24, 1O.?-25, 10.33-11.23. 
54 This pattern also occurs in 3.44-4.8. 
556.56-7.10 and 12.6-19 could be said to display either of these patterns. 
56 6.2-46, 7.15-8.30, 8.30-9.1. 
57 This section (5.9-54) has a significantly higher concentration of gaps than any 

other. 20% of the lines in this section have a gap (9 out of 45) compared with less than 
100/1! (39 out of 419) of the lines in the other sections which have gaps and about 8% 
(48 out of 606) of the lines in the entire Code. 



GAGARIN, MICHAEL, The Organization of the Gortyn Law Code , Greek, Roman and 
Byzantine Studies, 23:2 (1982:Summer) p.129 

144 THE ORGANIZATION OF THE GORTYN LAW CODE 

(I. Cret. IV 1-40) gaps do not occur except at the end of a line, prob­
ably because each separately enacted provision was inscribed on a 
new line beginning from the right. In the inscriptions of the later 
period (41-140) there are many gaps followed by a connective (nor­
mally aE), and in almost all cases a close connection between the 
provisions on either side of it is either likely or (in the absence of a 
clear context) possible.58 Consider, for example, 47, which, if not 
complete (see supra n.26), at least preserves a full and clear context. 
The first ten lines contain provisions concerning a slave who commits 
an offense while mortgaged to another man. After a gap there follow 
provisions concerning an injury to or the death or disappearance of a 
slave mortgaged to another man (lines 10-33). The pattern here is 
similar to that in 3.17-37 of the Great Code and can be similarly 
explained on historical grounds. 

Only in one other inscription are the gaps puzzling: I. Cret. IV 80, 
which contains various regulations concerning relations between Gor­
tyn and the Rhizenians. The presence of three gaps in fifteen lines59 

seems to indicate the accumulation of separately enacted provisions, 
and the extremely diverse content of the regulations supports this 
view. One gap occurs in the middle of a sentence (11-12):60 "If 
(certain officials) do not exact (the fine), the elders who exact them 
shall be immune (from prosecution)~ [gap] the things written, but 
not others. "61 The words after the gap indicate that the immunity is 
granted only as long as they exact "the things written," and it is 
possible that the main provision was first enacted by itself and that 
the restricting phrase was enacted later, perhaps after someone had 
misused his immunity. On the other hand, it is also possible (and 
prima facie perhaps more likely) that this treaty was a single enact­
ment and that the gaps in it are simply anomalous.62 

With few exceptions, therefore, the evidence of the gaps presents a 
consistent picture, which is most easily accounted for by a historical 
hypothesis. The Gortynians in the sixth and fifth centuries enacted, 
inscribed, and reinscribed a large number of different laws, and it is 

58 Gaps (without asyndeton) occur in 41, 42, 45, 47, 80, 83, 84, and 91. I would 
reject the gap postulated by Guarducci in her restoration at 81.22. 

59 There is a fourth gap before the asyndeton in line 1; see supra n.27. 
60 C( 9.15 of the Great Code (supra n.50). 
61 al BE Ka ~ 7TpaMovTt, TO~ 7TPEtytUTOV<; TOVTOV<; 7TpaMovTa<; a7TaTov f/-LEV [vac.l 

Ttl E-ypa/-LJLEV', aA.A.a BE JJi. 
62 There is no parallel for the first gap in line 12. The other two gaps Oines 4, 12) 

coincide with distinct changes in subject matter. Though the original stone does not 
survive, Halbherr's drawing (AlA 1 [1897] 206) does not suggest any physical reason 
for the gaps. 



GAGARIN, MICHAEL, The Organization of the Gortyn Law Code , Greek, Roman and 
Byzantine Studies, 23:2 (1982:Summer) p.129 

MICHAEL GAGARIN 145 

reasonable to understand the organization of these laws as at least 
partly determined by this activity. Once enacted, laws were frequently 
amended, but the amendments were added to the end of the earlier 
laws~ even when rein scribing laws and amendments, the Gortynians 
rarely attempted to integrate the provisions into a more coherent 
whole. 

III. Conclusion 

I have presented this examination of certain details of the organiza­
tion of the Great Code in the hope that it may help us see more 
clearly the nature of Gortynian laws. From at least the early sixth 
century the Gortynians enacted many laws on a variety of matters, 
particularly concerning the family and property. Laws appear to have 
been enacted as the need arose, i. e., as disputes concerning specific 
matters needed to be settled. After a period of time the Gortynians 
reinscribed many of these laws in more compact columns, and at the 
same time reorganized them to some extent. They still adhered 
closely to the wording of the original laws and only rarely did they 
alter groups of provisions to make them more systematic or orderly, 
though they did bring amendments together with earlier provisions 
into large sections and made an effort to assemble together related 
sections. 

This historical process suggests an attitude toward law somewhat 
different from our own. The Gortynians did not systematically revise 
their laws, perhaps out of a more respectful attitude toward earlier 
laws. That they did reinscribe and occasionally reorganize their laws, 
however, indicates that they understood some of the difficulties cre­
ated by the mere accumulation of separately enacted provisions and 
saw the need for greater order. But modern scholars must beware of 
expecting too systematic a presentation of, say, marriage laws at 
Gortyn. 

A similar warning holds true for Athenian law63 and probably for 
the rest of Greek law as wel1.64 When Athens in the last decade of 
the fifth century entrusted the republication of the city's laws to a 

63 c.r. Gagarin (supra n.l7) 21-29. 
64 Even modern law is not so systematic as is sometimes thought. The Constitution 

of the State of Texas, for example, is now burdened by scores of amendments added 
over the course of more than a century, many of them obsolete~ and yet it has resisted 
numerous attempts (most recently in 1974) at systematic revision, perhaps in part from 
a fear of change which the Gortynians would well have understood. 



GAGARIN, MICHAEL, The Organization of the Gortyn Law Code , Greek, Roman and 
Byzantine Studies, 23:2 (1982:Summer) p.129 

146 THE ORGANIZATION OF THE GORTYN LAW CODE 

body headed by Nicomachus, the result was not a revision of the 
laws but the gathering together and sorting out of a large number of 
separately enacted and in some cases widely scattered provisions.65 

Like the Gortynians the Athenians sought to give greater order to 
their laws, but they refrained from a systematic revision of individual 
laws. As a result the laws of both cities may show less consistency 
than some scholars seem to expect. It is vital for our understanding 
of Greek law that we be aware of the historical circumstances which 
produced the laws in the form in which they have survived for us. 
The details of the Great Code discussed above have, I hope, shed a 
small ray of light on the workings of this historical process.66 

THE UNIVERSITY OF TEXAS, AUSTIN 

May, 1982 

65 See R. S. Stroud, Drakon's Law on Homicide (Berkeley 1968), esp. 24-25. 
66 I am grateful to Professors Henry Immerwahr at the American School of Classical 

Studies in Athens and John Sakellarakis in Iraklion for facilitating my first-hand inspec­
tion of the inscriptions at Gortyn. I would also like to thank an anonymous referee for 
some helpful comments on an earlier version of this paper.