A TYPOLOGY OF ROMAN LOCKS AND KEYS
__________________
A Thesis
Presented to
Dr. Thomas Davis
Southwestern Baptist Theological Seminary
__________________
In Partial Fulfillment
of the Requirements for MAABS 5010
__________________
by
Tommas D. Pace
April 10th, 2014
Copyright © 2014 Tommas Dean Pace
All rights reserved. Southwestern Baptist Theological Seminary has permission to
reproduce and disseminate this document in any form by any means for purposes chosen
by the Seminary, including, without limitation, preservation or instruction.
APPROVAL SHEET
A TYPOLOGY OF ROMAN LOCKS AND KEYS
Tommas Dean Pace
__________________________________________________________________
Thomas W. Davis, Professor of Archaeology and Bib. Backgrounds, Supervisor
__________________________________________________________________
Steven Ortiz, Professor of Archaeology and Bib. Backgrounds, Second Reader
__________________________________________________________________
Mark Taylor, Professor of New Testament
Date______________________________
To my friends and family,
they who rally to my cry:
Gottes Freunde,
aller Welt Feinde.
ABSTRACT
A TYPOLOGY OF ROMAN LOCKS AND KEYS
This thesis addresses a gap in materials studies within archaeological research by
developing a functional typology of Roman lock and key forms for analytical and topical
research.
Section 1 introduces the reader to the background of the research, as well as the
author’s approach in chapters 1 and 2. A brief synopsis of the material’s literal and
cultural background is provided in chapter 3.
Section 2 develops the physical typology via examination of the available
materials to outline the various lock and key forms as they appear. Chapter 4 examines
the anatomy of locks and keys to establish terminology and address ancient lock
operation. Chapters 5, 6, and 7 detail the various forms and diagnostics which comprise
the typology.
Section 3 examines the materials in a physical context to establish
precedence for continued work in Chapter 8. The typology is concluded and listed are
avenues for future research.
Tommas D. Pace, M.A.
Advisor, Thomas Davis, Ph.D.
School of Theology
Southwestern Baptist Theological Seminary, 2014
TABLE OF CONTENTS
Page
LIST OF FIGURES .............................................................................................................x
PREFACE ......................................................................................................................... xii
Chapter
1. INTRODUCTION ................................................................................................1
Approach.....................................................................................................3
Previous Research .......................................................................................6
Conclusion ..................................................................................................6
2. THE FIRST LOCKS .............................................................................................9
Mesopotamian and Egyptian Locks ............................................................9
Early Tumbler Locks ................................................................................10
Shaft Keys .................................................................................................11
Early Tumbler Locks in Literature ...........................................................13
The Homeric Lock ....................................................................................14
The Egyptian Cylinder ..............................................................................18
3. SOCIAL CONTEXT ..........................................................................................21
Symbols of Power and Authority .............................................................21
The Cult of Portunus .................................................................................25
Semantics ..................................................................................................28
Roman Law and Status .............................................................................29
The Twelve Tables ...................................................................................29
The Status of the Roman Wife..................................................................31
Criminality ................................................................................................33
Locking and Sealing .................................................................................34
Keys and Eastern Texts ............................................................................36
4. LOCK AND KEY ANATOMY .........................................................................39
Lock Anatomy ..........................................................................................40
The Mount ................................................................................................40
Lockplate and Keyway .............................................................................40
Deadbolt ....................................................................................................43
Hasp ..........................................................................................................43
The Casing ................................................................................................44
Internal Components .................................................................................45
Key Anatomy ............................................................................................47
The Handle................................................................................................47
Hoop .........................................................................................................47
Flange or Bezel .........................................................................................48
Grip and Neck ...........................................................................................49
The Shank .................................................................................................50
Flat or Barrel .............................................................................................50
The Bit ......................................................................................................51
Tines .........................................................................................................51
Perforations, Grooves, and Channels........................................................51
Platform ....................................................................................................52
5. THE TUMBLER FORM ....................................................................................57
Tumbler Attributes....................................................................................57
One-Handed Tumbler Lock ......................................................................58
Two-Handed Tumbler Lock .....................................................................59
Key Forms for Tumbler Locks .................................................................60
The Laconian Form...................................................................................61
The Slide-Key Form .................................................................................63
The Lift Form ...........................................................................................64
Ring Keys for the Tumbler Lock ..............................................................65
6. THE ROTARY FORM .......................................................................................72
Rotary Locks .............................................................................................72
Warded Lockplate .....................................................................................73
Warded Locks ...........................................................................................74
Rotary Keys ..............................................................................................75
Grooved Rotary Key .................................................................................76
Perforated Rotary Key ..............................................................................77
Rotary Ring Key .......................................................................................78
7. THE PADLOCK .................................................................................................84
The Padlock Form.....................................................................................85
Casing .......................................................................................................85
Retainer and Barb .....................................................................................86
Compression Key Form ............................................................................87
Rotary Key Form ......................................................................................88
8. PHYSICAL CONTEXT .....................................................................................93
Chests and Cabinets ..................................................................................93
Doors and Gates ........................................................................................94
Keys to the City ........................................................................................95
Guilds ........................................................................................................96
Anthropomorphism and Zoomorphism ....................................................97
Zoomorphic Keys and Locks ....................................................................98
Anthropomorphic Keys and Locks ...........................................................98
Deity .........................................................................................................99
9. SUMMARY AND CONCLUSION .................................................................106
Summary .................................................................................................106
Conclusion ..............................................................................................108
BIBLIOGRAPHY ................................................................................................. 110
LIST OF FIGURES
Figure
Page
1. Replica of Homeric Key ...................................................................................... 20
2. Hecate with Laconian Key ................................................................................... 39
3. Early Roman Lockplate ....................................................................................... 54
4. Roman Deadbolt .................................................................................................. 54
5. Roman Tumbler Lock (Reconstruction) .............................................................. 55
6. Rotary Key ........................................................................................................... 55
7. Rotary Key with Greek Inscription ...................................................................... 56
8. Roman Key Handle .............................................................................................. 67
9. Crouching Figure Wielding a Laconian Key ....................................................... 67
10. Late Roman Slide Key ......................................................................................... 68
11. Roman Slide Key ................................................................................................. 68
12. Early Byzantine Ring Key for Tumbler Lock ...................................................... 69
13. Roman Slide Key ................................................................................................. 69
14. Roman Slide Key with Deadbolt ......................................................................... 70
15. Roman Lift Key ................................................................................................... 70
16. Roman Lockplates and Reconstruction ............................................................... 71
17. Roman Slide Key/Rotary Key ............................................................................. 80
18. Early Byzantine Rotary Key ................................................................................ 80
19. Early Byzantine Rotary Key with Grooved Bit ................................................... 81
20. Middle Byzantine Rotary Key ............................................................................. 81
x
21. Rotary Key ........................................................................................................... 82
22. Rotary Ring Key .................................................................................................. 82
23. Roman Ring Key.................................................................................................. 83
24. Barrel-Shaped Padlock......................................................................................... 90
25. Byzantine Padlock in the form of a Bull .............................................................. 90
26. Anthropomorphic Padlocks ................................................................................. 91
27. Rotary Key with Padlock ..................................................................................... 91
28. Reconstructed Roman Padlock ............................................................................ 92
29. Stone Chest inscribed with Tumbler Lockplate Form ....................................... 100
30. Ram Head Key Handle ...................................................................................... 100
31. Horse Head Key Handle .................................................................................... 101
32. Ram Head Key Handle ...................................................................................... 101
33. Key Handle in the form of a Lion ...................................................................... 102
34. Bounding Lioness Key Handle .......................................................................... 102
35. Bounding Lion Key Handle ............................................................................... 103
36. Key Handle in the Shape of an Arm, Clutching an Object ................................ 103
37. Key Handle in the Shape of a Human Head ...................................................... 104
38. Key Handle of an Orator .................................................................................... 104
39. Middle Byzantine Spring Padlock ..................................................................... 105
xi
PREFACE
No serious academic can say with integrity that their research is solely their own.
All advancement in the pursuit of scientific and social inquiry is accomplished through
the communitas of those around the researcher. This could not be more accurate in my
case. Though the provisional information of this thesis may be of benefit to future
researchers in the arena of materials studies, it could not have been completed without the
love and support of family and friends.
Dr. Tom Davis first insisted I approach this subject while working on the island of
Cyprus. I replied with an enthusiastic yes, which soon turned to the arduous fear that
such a work would bear no benefit to the field of Archaeology. It was not until we
stumbled upon a Roman key mislabeled as “medical instrument” on display in a museum
that I began to understand the need and efficacy for such a typology. I owe gratitude to
Dr. Davis, as well as Dr. Steven Ortiz, for their support in this work as reviewers of my
research.
I am indebted to my colleagues above all, for their advice and input as I
researched and formed the typology. Stephen Humphreys pointed out several
idiosyncrasies of Roman culture which were beneficial to my research. Andrew Johnson
originally pointed me to the deity and background of Portunus. Erin Daughters gave
editorial advice and suffered the resentful task of reading through various edits for style,
though any faults found within this thesis are not hers, but my own. Each of my
xii
colleagues, many of which are not listed, gave advice and input, but even more so, they
gave support and strength in an environment that is lacking in such.
But none sacrificed more for this work than my wife, Stephanie. Countless days
were spent when she would arrive home to stacks of papers and books cluttering our
living room. She stood by me amidst the pressure and gave up opportunities and
obligations to support my bad habit of Archaeology. She encouraged me and lifted my
spirit when I was exhausted and stretched beyond my limitations, to push me forward to
the goal.
Tommas D. Pace
Fort Worth, Texas
April, 2014
xiii
CHAPTER 1
INTRODUCTION
For much of the latter half of the 20th century, locks and keys have been placed on
the wayside of collections and exhibitions, likely due to their dubious technological
nature. Very few individuals know and understand how these small technological
marvels worked in the ancient world, just as very few understand how our modern locks
and keys work on a mechanical level today. The nature of security relies on a veiled
knowledge of the mechanics. Only the crafter or the patron knew the subtle intricacies of
the lock. When researchers examine artifacts of any kind, they are subconsciously drawn
to the form of the material in question and immediately evaluate the artifacts use based
on those typological features. This thesis will seek to analyze and evaluate the
physiological and social aspects of Roman locks and keys to establish a typology of the
forms currently evident from the archaeological record.
Cultural and social variances can be difficult to discern purely from an artifacts
form. In many instances, such as with ceramics and lithics, the form of the material may
be changed over time and hold a variety of uses. Ceramic vessels may be constructed to
store grain or oil, but can be broken and used as mortar in wall construction. Lithics will
often be reworked into different tools once they break or lose their edge. The intent of
archaeologists and anthropologists in forming typologies is to not only provide a
framework for understanding a specific artifact type, but to also understand any cultural
implications it holds. An anthropological understanding of the craftsman’s intent and the
1
2
patron’s intended use is the apex of any typology. In creating such a typology, it is
essential to address the methodology utilized for achieving such a goal.
How typologies are formed and what they can tell us is not an idle debate in
archaeology, as methodology and classification can cause division in final interpretation.
Likewise, anthropological evaluation can suffer if a proper typology is not constructed.
This can be seen in attempts to classify pottery. Attributing social change and evolution
of form can be a difficult task, especially since we know that by nature the archaeological
record will always be incomplete. Not everything has been excavated and not everything
has survived, leaving voids that we can clearly see and address by forming typologies.
This must always leave room for speculation since the record is incomplete. It is not a
pessimistic view however, since even a single sampling can sometimes speak so vividly
to a culture and an artifact’s production.
Lock and key production within the confines of Roman culture is an overlooked
medium. I chose this period of time for several reasons. The first is that the period of
Roman rule from before the Common Era to the rise of the Eastern Roman Empire at
Constantinople provides some of the most significant changes and additions to lock and
key technology. This technology disappears from use as the Byzantine period progresses,
and the technology is rediscovered and used again in the 19th century. The second reason
is that the Roman period exhibits the technology that its locks came from, utilizing the
oldest known form of complex lock in the Ancient Near East. This will grant researchers
a fluent background of the evolution of locks and keys. The final reason is that there is a
large enough sampling of data to warrant a typology and brief anthropological analysis
within the scope of a master’s thesis.
3
Approach
In formulating the typology, a combination of physical evaluation and
anthropological inquiry was incorporated into the corpus. Irving Rouse confers type to an
artifact as a combination of both artifact form and intent of the crafter and the user.
Rouse labels this evaluation in typology as an artifact’s mode.12
This essential
amalgamation of the physical form with the anthropological is not necessarily a more
efficient typology, but rather one that places anthropology at the core of the typology.
Any use of the term ‘mode’ in this thesis will be for the intent of applying the
aforementioned definition to an artifact. In many instances it can be difficult to discern
the mode of an artifact if it is incomplete or out of context, but in some instances we can
know the intent of the creator and the artifacts application. This typology utilized mode
throughout the formation of the work, as most keys revealed the lock for which the
instrument was used and conversely informed on the chora it likely protected.
Locks and keys are an effective medium for an anthro-centric typology.
Generally, all keys are meant to operate a specific locking mechanism as the mode of
each of these artifacts rarely strays from that intent. But keys can hold a duality in their
mode as they come in a variety of forms; zoomorphic, anthropomorphic, and decorative
keys and rings. These modes may grant a cultural evaluation of social standing. Outside
of the context of material culture, keys appear in text and depiction as important symbols
of social status, as well as in cultic practice, which will be discussed in the 3rd chapter.
1
Irving Rouse. “The Classification of Artifacts in Archaeology,” 25 American Antiquity
(1960): 313.
2
Dwight Read, Artifact Classification: A Conceptual and Methodological Approach. (Walnut
4
Locks rarely survive in the archaeological record and even when they are recovered, they
are often too corroded to be disassembled for examination of their construction and
intricacy. But a basic understanding of their form can be established as enough
information survives. Not every typology can rely on form and function equivalency, but
locks grant leeway in this matter as the function of the lock is entirely dependent on the
form of the key. Only the key crafted for a specific lock will operate said lock, and not
every key will even fit in every lock, only the one for which it was crafted.
The optimal approach for forming a typology would utilize social interview and
interaction to determine the exact mode of an object. Since we cannot speak directly to a
Roman blacksmith or his constituents regarding the construction and use of these
artifacts, this typology will rely on a predominantly etic evaluation contextualized in the
surviving emic information available through textual sources. An example of this
approach is seen in chapter 2. Very few archaeological samples survive in regards to the
Homeric lock and those that do survive have only a few indicators denoting them as such.
In this instance, surviving classical texts and modern examples have served to identify
these obscure artifacts. The provisional emic information is used within the typology
when applicable and if available.3
Individual chapters will approach each lock and key type with a deconstructive
________________________
Creek: Left Coast Press, 2007), 15.
3
At certain points, this research was unable to provide significant textual data to make cultural
inferences. Interpretation of the data is dependent on the observer in such instances and therefore, an
effective means of deconstructing the form is necessary. As such, Chris Caple’s (2008) work on
approaches to material analysis; deconstruction, evaluation, and methodology in Objects: Reluctant
Witnesses to the Past has provided the means to do so. The research and analysis to break down each
object used prior to the writing of this typology was performed according to Caple’s FOCUS format in
Chapter 1 of his book. His approach allows for a cultural evaluation based on an analysis of the
construction, use, loss, and reclamation of an artifact and has been utilized to form this thesis.
5
approach that classifies each type based on the materials’ function, dimensions, attributes,
and composition. The composition of each key is indicative of the materials used, and
therefore a window into the objects construction. The physical attributes of each key
may relay anthropological information in instances where decoration is used. There is no
delineating factor that differentiates a key’s use for a lock on a door versus a small chest
or cabinet. The dimensions of each key will acknowledge the size of the lock for which it
was used and subsequently help identify the possible scenarios in which that lock was
used.
The physiology of a key informs the investigator to the care of the artisan in its
crafting. But more importantly, combining emic and etic approaches through Rouse and
Caple, the physiology of the key tells us the type of lock for which it was most likely
used. Weight, size, and shape may not always be indicative of a key’s chora, nor the
intent the artisan placed on its construction, but it will speak to its utility. Likewise, the
level of sophisticated decoration or relative crudeness may provide subtle hints to the
utility of the key, but will certainly reveal the care taken in the key’s creation on part of
the craftsman. It is for these reasons that a duality of approach is necessary in forming
this particular typology. In many instances I was not afforded the data regarding the
background of many keys; their provenance, origin, or context. While that information is
beneficial for a complete understanding of locks and keys, by utilizing a deconstructive
approach, I am able to provide an efficacious typology that is loyal to the information
available.
The terminology used in establishing the anatomy is synonymous with
contemporary locks and keys. This was done for the sake of continuity, but also for
6
modern comparative analysis. The terms are based on both ancient and modern lock and
key labels, with some alterations due to archaic vocabulary and misnomers. Assimilated
terminology from ancient sources has been combined with the advent of the modern lock,
post Industrial Revolution.4 While the materials used for the construction of locks and
keys has changed significantly over two thousand years, the form and operation has not.
Most ancient locks and keys share attributes with contemporaneous forms in one part or
another. Much of the terminology in this approach would be recognizable for any
persons acquainted with the modern forms and can be used for comparison.
Previous Research
This thesis provides relatively new information regarding a specific segment in
material studies. Though locks and keys have been examined in the past, primarily
through the work of Dr. Gary Vikan, no formal typology has ever been attempted. Dr.
Vikan began examining materials specifically from the Byzantine period, which bears
resemblance to the forms and functions of the earlier Roman forms and is found in his
published work with John Nesbitt, Security in Byzantium: Locking, Sealing, Weighing.
Dr. Vikan had begun a critical examination of Byzantine locks and keys within the Menil
Collection of Houston, Texas, but his research was part of a project that was never
published. Dr. Vikan’s work was made accessible to me for this thesis and has proven
beneficial for an understanding of the evolution of Roman lock and key forms into the
Byzantine period.
Conclusion
4
Vincent Eras. Locks and Keys throughout the Ages. (New York: Lip’s Safe and Lock
Manufacturing Company, 1957), 12-16.
7
The reader of this thesis needs to understand above all else that this is not a
quantitative typology. I have not gathered and counted every known Roman or
Byzantine lock and key, due to issues of accessibility. A quantitative analysis is beyond
the scope of a master’s thesis that is inclusive of a typology and overview of the
mechanical functions of locks, and not every key and lock is published nor even known
in every collection. But a quantitative analysis for the purpose of building a corpus of the
material can serve the understanding of these artifacts and is a strong candidate for future
research in this area.
Most locks were also composed primarily of iron, an element which does not
often survive the archaeological record. Those locks which survive are incapable of
disassembly; being in a delicate state of decay their preservation would be compromised.
This does not pose a problem for this typology. The form of a key always follows its
function as each key is crafted for a specific lock. The general principles of lock-type are
consistent throughout the ancient world and are evidenced by gradual technological
change. The keys themselves provide the crux for understanding and shaping lock
typology as you will see in subsequent chapters. The form and principle of each lock
type will be broken down in an individual chapter, along with their artifact counterpart,
the key. While the keys themselves will vary in form, design, and intricacy, their
functional principles reveal the lock for which they are used.
My hope in completing this typology is to provide the reader with an
understanding of locks and keys within the context of the socio-cultural atmosphere in
which they were created. In providing the forms and identifying markers of locks and
8
keys, it can serve as a manual for identification in collection and in the field, to grant
these materials an anthropological place in both the museum and excavation.
9
CHAPTER 2
THE FIRST LOCKS
The origin of locks and keys remains ubiquitous in the archaeological record.
Undoubtedly, man learned to secure his belongings not long after the agricultural
revolution, as sedentarianism led to urbanization. Pliny the Elder attributes the origin of
locking mechanisms to Theodorus of Samos, though his understanding of contemporary
technology is clearly shaped by a Greco-Roman worldview.5 Such devices, however,
have been found in use long before the mechanical creations found in the Roman period.
Mesopotamian and Egyptian Locks
Locks, even those as simple as a wooden bolt securing a door, survive on rare
occasions early on in the archaeological record. This is evidenced in some of the first
locks and keys that appear in Mesopotamia and Egypt. The construction of these locks
were typically from wood, but locks and keys made from bronze and iron were utilized
after the Chalcolithic as well.6 It is from this design that the modern lock takes its form,
as well as the majority of locks discovered dating to the period of Roman rule in the
Mediterranean. Because Roman technology evolved from these forms, it is important to
acknowledge and understand their type and use, as they are utilized in all periods.
5
Pliny the Elder Naturalis Historia. 7.198
6
John Curtis and Matthew Ponting. An Examination of Late Assyrian Metalwork: With Special
Reference to Nimrud. (Oxford: Oxbow Books, 2013), 58.
10
The earliest evidence of locks and keys come from Khorsabad, Iraq at the palace
built by Sargon II.78 Examples synonymous with this type also survive at Karanis, Egypt.
These date to about the period of Ramses II. Both have examples that are composed of
wood and are distinguished not by their construction, but by the technology of their use.
These first locks were ingeniously constructed as tumbler locks and were designed to
obfuscate any attempt to open a passage, save by possession of the appropriate key. The
composition of the locks and keys is typically wood and though organic materials rarely
survive the archaeological record, some artifacts are not subject to decay in a desert
climate. The security mechanisms and doorways found in Egypt are constructed of
acacia wood, though some are made from a weaker palm wood.9 They are preserved due
to the low moisture climate and buried in anaerobic conditions, preserving them to the
modern day.
Early Tumbler Locks
The first known lock and key technology that is exemplified in these contexts is
the tumbler lock. The function of a tumbler lock is characterized by the use of tumbler
pegs housed within a casing. The pegs have a recess within the deadbolt that keeps the
bolt from moving of its own volition and can only be operated with the insertion of the
appropriate key. The application of ‘tumbler’ to this lock and similar locks of the Roman
period is based on their synonymous function with modern security.10 Once a deadbolt is
7
Ibid.
8
Peter James and Nick Thorpe. Ancient Inventions. (New York: Ballantine Books, 1994), 471.
9
Elinor M. Husselman. Karanis Excavations of the University of Michigan in Egypt 19281935: Topography and Architecture. (Ann Arbor: The University of Michigan Press, 1979), 40-41.
10
Eras, 16.
11
thrown into position and the key is removed from the housing, the pegs ‘tumble’ down
their channels into a drilled fitting within the deadbolt. Modern tumbler locks utilize this
same concept with pins of various lengths. Later Roman technology includes the use of
metal springs, but the earliest locks relied on gravity to pull the tumblers down into
position.
Only with the proper key inserted into the lock could the operator correctly
manipulate the wooden tumblers out of the deadbolt. With the tumblers pushed out of
position, the operator can pull the bolt from its recess, unlocking the door and thereby
opening it. Tumblers are carved in various sizes and lengths to render the lock unique
from any other. As such, the keys constructed for the lock are constructed in such a
manner to appropriately manipulate the tumblers. The position of locks on the doorways
and passages of Karanis do not appear at a designated height. In several instances, locks
appear to have been installed at the base of the door, whereas others have been installed
around waist-level.11 At Khirbet Deir Sam’an, defensive structures relied on heavy beams
for bolts, secured in the threshold vertically, as opposed to the frame, as is typical with
general passageways.12
The constructions of such devices vary in size, dependent upon the passage or
chora the lock is securing. Large locks and keys are required to secure larger beams and
passageways, though it appears that most defensive structures rely on less complex
methods of securing from unwanted entry as is evidenced by the large number of
11
For a visual comparison, see Husselman. Compare plates 42, 46, 48, 50, and 53 for variations
in height for the deadbolt recess.
12
Noga Haimovich-Carmin, “Churches and Monasteries in Samaria and Northern Judea ,” In
12
passageways that are secured by a simple beam of wood.13 Remnants of simple locks,
much akin to the ingenuity of the Homeric concept are retained at Khorsabad, which
utilizes beams set inside copper rings to shore up doors and gates, likely for internal
security for the structures at night, and possibly to defend during siege.14
Shaft Keys
The unique form of key used for the earliest tumbler locks is markedly different
from its later Roman counterparts. The early keys found at Karanis display distinctive
qualities in their use that differentiate them from the slide-keys that are discussed in
chapter 5. Typologically, the shaft key maintains the same characteristics from which
later keys will be shaped. Shaft keys have tines that rest on a platform, or are integrated
with the platform, that manipulate the tumblers within a lock. The surviving versions of
Karanis have linear tines of various shapes and heights to correspond with the appropriate
lock. The distinguishing feature of shaft keys is in the body of the key itself.
The body of the shaft key is elongated, often made from a single piece of wood.
Though one of the shaft keys displays a slight elbow in its form, the placement of the
tines and platform as parallel with the shaft corresponds with the use of the key. The
term ‘shaft’ is used for the key’s utility. The key and the operator’s arm are placed
through a port in the door and the bit of the key is inserted into a hollowed-out recess, or
shaft, within the deadbolt. The design of shaft-keys allows for easier passage of the
________________________
Christians and Christianity, Vol. III. (Jerusalem: Israel Antiquities Authority, 2012), 17, 43, 46.
13
Husselman, 40-41.
14
Curtis and Ponting, 56-57.
13
object and effective operation of the locking device. The form is comparable to the slide
keys that appear throughout the Roman period, but differ in their mode. The tumblers of
Roman locks perforate the deadbolt entirely, whereas the earliest tumblers do not.
Roman keys found in the archaeological record do not follow the classification of
shaft keys, but this does not mean they were not used in Roman culture. In
Metamorphoses a criminal break-in is described that indicates the technology of the lock
which bars their path.
“We lost no time, and by nightfall we were ready at his front doors. We decided not
to remove or force them apart or break them down… And so our noble standardbearer Lamachus, with all the confidence of his tried valour, gradually slipped his
hand through the hole for inserting the key and attempted to dislodge the bolt.”15
The description leads one to believe that Lamachus put his hand through the keyhole, but this is unlikely, as the majority of Roman locks have small L-shaped keyways
far too small for the insertion of a human hand. The passage does serve to describe the
common assembly of early tumbler locks which utilized shaft keys or even the twohanded tumbler lock. These early locks have a port access to allow an entire human hand
to be inserted for the passage of the key into the deadbolt. Manipulation of the tumblers
would still prove difficult without the key, but such a locking mechanism fits the
description of Apuleius’ text. Though iron and bronze are prominent in use amongst the
Romans, wooden locks and keys are likely constructed by the poor, or where simple
locks and keys are necessitated.16
Early Tumbler Locks in Literature
15
Apuleius Metamorphoses. 4:10.
16
Augustine of Hippo De Doctrina Christiana. 4.11.26.
14
Ancient locks and keys constructed before the advent of copper alloys and iron
rarely survive, but indications of their construction and use exist within the literary
tradition of the Hebrew Scriptures as well as ancient Near Eastern texts. The story of the
assassin Ehud, a judge of Israel, provides a subtle description in the use of these ancient
locks.
“Then Ehud went out into the vestibule and shut the doors of the roof chamber
behind him, and locked them. When he had gone out, his servants came and
looked, and behold, the doors of the roof chamber were locked; and they said, ‘He
is only relieving himself in the cool room.’ They waited until they became anxious;
but behold, he did not open the doors of the roof chamber. Therefore they took the
key and opened them, and behold, their master had fallen to the floor dead.”17
When Ehud departs the room, he prolongs his chance for escape by locking the
door behind him to confound the king’s guard. This can be done without a key with
tumbler locks that utilize shaft keys. Ehud would only need to reach his arm through the
port constructed in the door and pull the deadbolt into position by hand. The tumblers
would then fall into position, locking the door. The deadbolts for Roman locks function
in much the same way. The deadbolt can be pushed into a locking position and the
tumblers will fall into place. To understand this concept, see the reconstruction of the
tumbler lock in Figure 5 found at the end of chapter 4.
Though the Hebrew Scriptures give subtle hints to the form and function of locks
and keys pre-dating the early Roman republic, the literature and inscriptions found from
Mesopotamia affirm such devices as the prominent form amongst Sumerian culture. An
extensive repertoire describing locks and keys in addition to their use survives from these
texts.
Armas Salonen sought primarily to establish a philology of Sumerian and
17
Judges 3:23-25. I have used the NASB translation throughout this thesis.
15
Akkadian texts as the core for his research, but his work, Die Tueren des Alten
Mesopotamien, forms a comprehensive lexicon of the subject.18 The terminology of the
texts reveals that similar parts and concepts were used in the construction of ancient shaft
keys and locks. The architecture of doors is comparable, pivoting on shoes and are likely
secured in the same manner as the doors preserved from Karanis, Egypt.1920 Locks are
described as using pins and deadbolts to secure the devices, accessible via shaft.21 The
anatomy of the key has a bit to manipulate the lock.2223
The Homeric Lock
The Homeric Lock in Literature
Though a typology of locks and keys can be formed from the relative data found
in the archaeological record, there is no doubt that a variety of these constructions have
been used in various cultures throughout the world that do not match the established
technology found in the Roman Period. Where a typological lock or key cannot be
manufactured, or a patron cannot afford a crafted lock, human ingenuity is substituted.
Evidence of locks and keys that do not fall into the classification of tumbler, padlock, or
rotary are found in descriptors in both ancient text and archaeology.
18
Armas Salonen. Die Tueren des Alten Mesopotamien: Eine Lexikalische und
Kulturgeschichtliche Untersuchung. (Helsinki: Suomalaisen Kirjallisuuden Kirijapaino, 1961), 4.
19
See Plates IV, V, VI, and VII for Comparison.
20
Salonen, 170-173.
21
Ibid., 75-78.
22
Ibid., 87.
23
In several instances, the function of the aforementioned devices is realized through Sumerian
allegory and metaphor, revealing that such technology had a significant socio-cultural role in everyday
language and use. See: Salonen, pp. 74-92, “Riegel, Schloesser, Schluessel und Handgriffe” for examples.
16
One such case is in Homer’s Odyssey. When Penelope leaves the room of the
suitors to retrieve her husband’s bow, she must operate a lock to do so. The text
describes both key and lock in fair detail:
“Up the steep stairs to her room she climbed
And grasped in a steady hand the curved key—
Fine copper, with ivory haft attached—
And then with her chamber-women made her way
To a hidden storeroom, far in the palace depths.”24
“The lustrous queen soon reached the hidden vault
And stopped at the oaken doorsill, work an expert
Sanded smooth and trued to the line some years ago,
Planting the doorjambs snugly, hanging shining doors.
At once she loosed the thong from around its hook,
Inserted the key and aimed straight and true,
Shot back the bolts—and the rasping doors groaned
As loud as a bull will bellow, champing grass at pasture.
So as the key went home those handsome double doors
Rang out now and sprang wide before her.”25
The passage describes noteworthy details in regards to ancient locks and keys,
though the description of the key as copper can be misleading. Smiths did not typically
utilize native copper in the construction of keys or locks, due to its ductile nature.
Copper, being malleable if not alloyed with arsenic, antimony, or tin, will quickly bend
and reshape after a short period of use.26
It is plausible that Greek does not differentiate between copper and copper-alloys
in description. While the properties of copper in the absence of a reagent to strengthen it
do not discredit the use of copper, it does help to explain the form of the lock in the
24
Homer The Odyssey. 21.6-10.
25
Ibid., 21.50-59.
26
John F. Healy. Mining and Metallurgy in the Greek and Roman World. (London: Thames
and Hudson, 1978). 209-214.
17
passage. The key described would have been nothing more than an angled copper rod,
shaped and designed for that particular door. The locked closet would have borne a hole
or passage through which the copper rod could pass, and with an appropriate
maneuvering of the hand and wrist, a thrust of the key would push the deadbolt into
retraction. The bolt itself would have a notch or protrusion for insertion of the key. It is
likely that the rod and bolt would have both been fashioned to complement one another in
use. A replica of a broken Homeric key is retained in the archives of the British
Museum. (See Fig. 1) An attic ware vase at the Berlin Altes Museum depicts a
Hellenistic woman using such a key, but unfortunately, the vase was lost in the Second
World War.2728 The woman is shown using the key, inserting it near the top of the door
for operation. The handle appears to be identical to the replica in Figure 1.
Since the deadbolt for the Homeric style of lock does not operate on a complex
system of tines or levers, only a simple key is necessary for use. A copper rod can be
reworked if bent and would serve this function dutifully. These minimalistic
constructions are utilized throughout the ancient world and occurred in situations where a
complicated lock was not required. For the passage from the Odyssey, the simple lock is
found on the door to a strongroom. In other circumstances, variations of the same lock
would likely have been found throughout the home. Though a more complex version
may have been used on any primary entrance, it was not always the case. The temple of
Artemis at Lusoi appears to have retained a Homeric lock on the entrance or within the
27
Staatliche Museen zu Berlin, Antikensammlung. Foto: Archiv, Inventar Nummer F 2382.
[Kriegsverlust].
28
A surviving photograph is on file at the museum, but only partially shows the key in
operation. A better picture still survives in copies of Die Technik Des Altertums, by Albert Neuburger.
18
temple, as a key survives with an inscription designating it as such survives from the 5th
Century BCE.29
The Egyptian Cylinder
Arid regions of the Ancient Near East are known for utilizing locks in much the
same way as the Homeric key, but with variation. These locks range from the simple to
the complex and are even used in contemporary societies. The discovery of various
wooden cylinders in Egyptian tombs puzzled archaeologists for quite some time, leading
to speculation of their use. But in a bout of unintentional ethno-archaeology, it was the
discovery of Homeric style devices which were still in use amongst Ethiopian villages
that allowed Sir Flinders Petrie an understanding of the function of such artifacts.30
Ethiopian villagers were utilizing an ingenious system of hollowed cylinders which could
be attached to ropes and inserted through a key hole. Upon pulling the rope, the cylinder
would be forced against the interior deadbolt to push it out of a locked position.
Locking the door was done in the same manner as the descriptions found in
another instance of Homer’s Odyssey when Telemakhos departs to his room and his
servant locks the door behind her.31 At first glance this text makes it seem as if
Telemakhos is being locked into a room from the outside, but it is not the case, as the
room likely has the Homeric lock installed to secure the room upon departure.32 The rope
29
Charles Singer, E.J. Holmyard, A.R. Hall, eds. A History of Technology, Vol. II. (Oxford:
Clarendon Press, 1954), 416.
30
James and Thorpe, 469.
31
Homer, 1.500-504.
32
Compare with 21.262-3 of the Odyssey, where Odysseus tells the serving women to lock
themselves into their own rooms for their safety.
19
or strap would be strung through the keyhole and knotted and could be locked by tugging
the rope, thereby also pulling the attached deadbolt. Unlocking the door was a simple
matter of stringing the pre-drilled cylinder with a second rope already connected to the
deadbolt. Once the two ropes are threaded through the cylinder key, the key is pushed
through the port of the door and the secondary rope tugged, pushing the deadbolt back
into the door frame and unlocking the door.33
33
Though such an example is not displayed in this thesis, a proper illustration outlining the
form and function of such devices is found in the entry, “Locks and Keys” within Ancient Inventions by
James and Thorpe, p. 170.
20
Figure 1: Replica of Homeric Key
Image AN374403001
Courtesy of the Trustees of the British Museum
21
CHAPTER 3
SOCIAL CONTEXT
Archaeology must appeal to the written record of ancient historians and orators to
garner a better understanding of locks and keys within their semantic and social context.
Though no ancient Roman writer speaks directly of the subject, examples of locks and
keys appear within the context of their use in a variety of early Roman writings. From
Cicero to the early Christian apologists, examples of locks and keys in use appear as
subconscious metaphors, often relating to another subject. The operation of the devices,
their strengths and weaknesses, as well as examples of criminality all exist in the context
of the ancient writers, orators, and playwrights. Locks and keys have a shared semantic
meaning with many of the modern uses of their terminology as allegory in both Greek
and Latin. Such writings relay passive anthropological information on the subject that
does not broach the archaeological record.
Symbols of Power and Authority
“I have Zeus behind me. Do we need to speak of that? I am the only god who
knows the keys to where his thunderbolts are locked.[…] So yield to my
persuasion and do not hurl the words of a reckless tongue against the land, that all
things bearing fruit will not prosper.”34
Athena’s words via the playwright Aeschylus echo a sense of power and authority
in a casual threat. In this regard, the goddess’ knowledge of the keys to unlocking Zeus’
34
Aeschylus Eumenides. 825-830.
22
power places her on par with the great god himself. Keys displayed in Greek art typically
follow the homeric form. The elongated bar with an extended bit which is bent at right
angles is typical of such keys. As was discussed in the previous chapter, this is the key
described in the Odyssey. Hecate is often displayed with either a torch or key in hand,
but it is shown to be of the slide-key type for a tumbler lock, with three tines
characteristic of the Laconian Form. (See Fig. 2). Her association with passageways and
entries is similar to the functional relationship of Portunes in Roman religion, but they
are not considered synonymous. Hecate was commonly worshiped in Greek culture for
good luck and fortune, but held a position of power that Portunes did not. The key that
Hecate holds is the key to the gates of Hades. This relays a symbol of power over death
and life. While the two share a commonality with transition, Hecate was greatly
associated with crossroads and the Underworld. Hecate’s power over life and death,
reflected by lock and key is the pinnacle transition of the body and soul into the afterlife.
In the context of the Christian New Testament, Jesus grants a symbolic power and
authority to St. Peter in Matthew 16:19, in which He verbally gives Peter the keys to the
Kingdom of Heaven. St. Peter is often displayed in medieval and modern art wielding a
key or set of keys. These keys are anachronistic, as they are typically Byzantine in form,
being of the rotary key taxonomy with cruciform perforations on the bit. The Peter of the
1st Century CE would have only been acquainted with slide-keys and perhaps some
simple rotary keys. The cruciform does not become a known symbol in Christian culture
until after Constantine.
The key bearer as a wielder of power was a consistent theme throughout the
ancient world and even persists through later periods. The Messianic passage of Isaiah
23
22:22 notes that the Messiah, ‘will be a key-bearer for the House of David’. The text
indicates that the key would be placed upon his shoulder. While this may be an allegory
for burden, the context dictates an attitude of authority and power. A large key would
also not be carried around the waist, but certainly could be hoisted onto a shoulder. The
wooden constructions which predate the Roman period would vary in size and the locks
and keys would also. The size of the key would correspond to the size or power of that
which it secures. The period of Isaiah is contemporaneous with Sargon, and the
technology would have shared similarities with the Mesopotamian-Egyptian locks and
keys discovered.35 The acknowledgment to power and authority would therefore not be
lost on the early readers of the passage.
According to Aulus Gellius, in his work Noctes Atticae, priests serving the temple
of Jupiter were not allowed to wear rings on their fingers that bore gems, but were
allowed to wear any other kind of ring.36 Depending on the interpretation of the text, item
iurare Dialem fas numquam est; item anulo uti nisi pervio cassoque fas non est, the
allowance of the priest to wear a ring that is perforated may indicate that key-rings were
worn by priests at this period. Being both common and utilitarian, key rings would likely
have been worn by priests to access the various securities stored within the temple. The
admonishment to not wear a ring with a gem seems to be a social indication of an
35
James and Thorpe, 471.
36
Gellius Noctes Atticae. 10,15.6.
24
estrangement from affluence, to prevent the priest from flaunting decadence to the
population.3738
As symbols of power and authority, keys signify a cultural understanding of
wealth and status. While most keys are constructed of bronze or iron, there exists one
known example of a key made from gold, which dates to 9th Century Byzantium.39 This
key also doubles as a signet ring with a monograph, a practice that became common in
the Byzantine world.40 The diary of Gonzales de Clavijo gives reference to a silver key
that is used in conjunction with sealed hasps.41 The Roman Republic and Empire
predating the rise of the Eastern Roman Empire has yet to yield an artifact of either
composition. The existence and survival of such an artifact is rare in any case, due to the
lustrous and coveted nature of gold amongst ancient robbers and the metal’s reuse. But
this does not imply that such compositional keys did not exist. On the contrary, gold
keys may have been common enough amongst the social elite as Augustine of Hippo
makes reference to such objects. “For of what service is a golden key, if it cannot open
what we want it to open?”42 Though Augustine is addressing the function of teaching in
this context, he is giving reference to a contemporary object. Clement of Alexandria
37
Ibid., 10,15.5-6.
38
Gellius is citing a work from his period (c. 125 C.E.) that does not survive in record; On the
Public Priests.
39
Gary Vikan and John Nesbitt. Security in Byzantium: Locking, Sealing, and Weighing.
(Washington D.C.: Dumbarton Oaks, 1980), 5.
40
Ibid., 10-28.
41
Ibid., 8.
42
Augustine, 4.11.26.
25
writes, “The Word permits them a finger-ring of gold. Not for ornament, but for sealing
things which are worth keeping safe in the house in the charge of housekeeping.43
The Cult of Portunus
Of the pagan Roman gods that envelop the pluralistic religion of the Republic and
Empire, Portunus appears as a titular deity for security and passageways. Historians
conflict on the facets of this minor deity, as classical texts allude to various personalities
and functions of the god. In the early days of the Republic, Portunus held a special place
in a festival of his honor.
On August 17th, on the Roman Kalend, citizens would celebrate Portunus by
sacrificing keys in an altar at the deity’s temple in Rome.44 This festival was known as
Portunalia and attracted a great number of those seeking good fortune and security in the
coming year. Before the ascension of Ianus as a favorite of the populace, Romans sought
the good will of Portunus for safety and security. Both Ianus and Portunus have served
as gods of passageways and transition, but it appears that the personality of Portunus
evolved as more and more Romans looked to Ianus throughout the Republic and then
Empire. It is argued that Portunus became associated with safeguarding travel by sea, as
well as ports and harbours. Indeed, in classical texts it appears that he becomes a favored
god of sailors. Virgil’s Aeneid gives credence for this postulation, as the favorable
conditions in which Cloanthus wins a naval race are revealed in his prayer for offerings
43
Clement of Alexandria Paedogogus. 3.11
44
Rudolf Haensch. “Inscriptions as Sources of Knowledge for Religions and Cults in the
Roman World of Imperial Times,” in A Companion to Roman Religion, (Malden: Blackwell Publishing,
2009), 179.
26
to the god who answers him.45 Cloanthus offers stipulation for assured victory, that the
god who answers the prayer will receive an altar upon which wine and choice bovine will
be sacrificed. Virgil writes of the answer to the prayer that it is Portunus who answers,
manifested by large hands that come out of the water and push the ship.46
There is no doubt that by this early in Roman history Portunus was seen as a deity
of the waters, but his shared similarity of a key as his symbol with Ianus and the general
relationship of keys and security with passageways implies that he has an origin of
association with safety that became synonymous with maritime travel. Cicero comments
on the origin of the names of the pantheon in De Natura Deorum, in which he notes that,
just as Neptune is derived from the latin nando for swimming, so is the name Portunus
from his relationship with ports.4748 Ports and harbors share a natural relationship with
security in the sense that they were safe locations for ships to dock and weather severe
storms and raging waters. They also served as points of destination and departure, a
thematic repose for a god of transitions.
Joachim Marquardt, one of the few scholars to comment on Portunus, attributes
the god to be a deity over warehouses.49 This may seem in contradiction to the maritime
nature we see in the Classical Latin texts, but etymologically it may be asserted. When
45
Virgil The Aeneid. 5.235-238.
46
Ibid., 5.241-242.
47
M. Tullius Cicero De Natura Deorum. 26.
48
See also: Varro De Lingua Latina. 6.19
49
Joachim Marquardt. Römische Staatsverwaltung, Vol. III. (New York: Arno Press, 1975),
327.
27
Cicero commented on the birth of the names of the gods, he attributed portu to the noun
associated with ports. But it also shares a fraternal relationship in linguistics to porta, the
noun characterized for doors and gates. Anthropologically, it can be asserted that ports
and passageways share a cultural heritage through linguistics that was acknowledged by
Portunus’ worshippers. The ubiquitous relationship between a god of the waters and his
characterization as a key-bearer reveals the emic understanding Roman citizens held of
the god.
Keys share the relationship of transition and security with ports, making them a
natural identifying attribute of Portunes and Ianus. Unfortunately, little information
survives on the cult of Portunus itself, and only the festival day of Portunalia is preserved
from antiquity. But it is known that the Temple constructed in Rome did retain its own
flamen for priestly duties involving the Temple and joint-worship structure of the
pantheon.50 This in turn posits that the flamen oversaw the festival of Portunalia in
addition to the sacrifice of keys that would take place. The keys that were cast into the
flames may have been facsimiles created by a blacksmith for the festival itself. Keys
were made individually or cast into molds to suit the locks for which they were
constructed. The possibility exists that either unusable or broken keys were sacrificed,
but it would have been unlikely for a Roman citizen to cast a working key into the fire
unless they had access to a duplicate. Such duplicates would therefore be bronze, as iron
was incapable of being cast into molds in the Classical era.
50
Joachim Marquardt and Theodor Mommsen. Handbuch der Römischen Alterthümer.
Sechster Band. (Leipzig: Verlag von S. Hirzel, 1885), 326-328.
28
Economically, metal was just as much a precious commodity in the Empire as it
was in any other period. Sacrifices of valuable objects are not uncommon in the Near
East and in the Roman world. Gold, silver, food and drink were often such sacrifices and
may have helped fund the temples in addition to regular tithes by devotees. A sacrifice of
keys would benefit the temple to Portunus on the festival day. Bronze was a particularly
precious metal and could certainly be re-smelted and cast into an ingot. Iron faced
difficulties in this regard, as it could not be smelted. But the structure of iron allows it to
be reworked or fused if heated well enough to other ferrous objects. Once Portunalia had
ended, the flamen may have gathered the sacrificed keys to be resold as bulk metal for the
funding of the temple.
Semantics
In Latin, clavis is used to denote a key, whereas sera or claustrum can denote a
lock. Likewise, the contemporaneous Greek uses κλεις for a key, but has several words
for the lock: μοχλός, θύρας, όχεύς.51 Clavis adulterae appears to be used to personify a
falsehood found in a response or answer, but it is also used quite literally as ‘adulterated
keys’ or ‘false keys’ in several instances where unlawful skeleton keys or duplicates are
made for criminal entry into a residence or gate. But such words are also used as
metaphor or allegory to accentuate or ascribe a purpose in Latin. Ovid uses the phrase,
clavigerum verbis adloquor ipse deum or, “key words of the gods”, in his prose to
emphasize a forthcoming statement.52 Euripides does the same in Medea, “May that man
51
M. Cary, ed., The Oxford Classical Dictionary [OCD], (Oxford: Clarendon Press, 1949), s.v.
“Keys and Locks,” by F.N.P.
52
Ovid Fasti. 1.228.
29
die unloved who cannot honor his friends, unlocking to them his honest mind.”53 It
appears that the use of the vocabulary in Latin and Greek held a syntactical versatility
that is still common in many modern languages.
False keys used for unlawful entry are mentioned by Ovid and Sallust.5455 The
translation in all instances is ambiguous as there is a stark contrast between a duplicate
key and a skeleton key. Undoubtedly, a duplicate could be made if the original is
present, but the clavis adulterinae do not always appear to be duplicates, but possible
devices of lock picking. Skeleton keys, if they are to be translated as such, are implicit of
their use within rotary locks. An individual can carry a variety of minimalistic keys
which can open the majority of rotary locks. But such keys cannot function with tumbler
technology. Unfortunately the translations dealing with physical ‘false keys’ are left
ambiguous, as a stark contrast exists between a duplicate key as a copy and a skeleton
key: which is essentially a lockpick. The purpose of a lockpick is to manipulate the
internal components within the casing to unlock the device without the use of the proper
key. Picking locks is a known practice amongst locksmiths and thieves of the modern
age and it is plausible that such a roguish craft exists within the Roman Republic as well.
Roman Law and Status
The Twelve Tables
53
Euripides Medea. 660-661.
54
Ovid Ars Amatoria. 3.643.
55
Sallust Bellum Jugurthinum. 12.10
30
The earliest known records of Roman law do not survive in antiquity. They do
survive however, in the literary sources of early Latin orators and historians. The original
Twelve Tables, or Twelve Tablets as they are sometimes called, refer to a series of
guidelines prescribed by the Roman governance in the early growth of Rome. The tablets
are purported to have been erected as bronze monuments, inscribed with simple language
so that citizens viewing them in public could know and understand the general standards
expected of the people.
Knowledge of these tablets appears to be widespread in the Roman Empire, as
Cicero , Pliny, Plutarch, and others reference them in orations even after their removal
during the sacking of Rome. The contents of the tablets and their inscriptions are lost,
save for passive citation in classical texts and literature. One tablet in particular appears
to have dealt with the social matters of marriage and divorce as both Cicero and Plutarch
address the topic, making mention of the role of keys on the subject. Marriage was
mostly transitory in the Roman world, capable of being enacted or ended at a whim on
part of the husband.56 The wife could dissolve a marriage under specific circumstances,
but only if his actions were socially egregious. Divorce could also take place for
household infractions involving keys.
Plutarch states that Romulus enacted divorce law with specific requisites for a
man to have permission to divorce his wife if she duplicates the household keys. For the
wife to do such was seen to be as heinous as poisoning the children, or being a
56
Matthew Kuefler. “The Marriage Revolution in Late Antiquity: The Theodosian Code and
Later Roman Marriage Law,” Journal of Family History 32 (2007): 355.
31
drunkard.57 But if the husband wished to divorce his wife for any other reason, she was
liable to receive half of his property in the divorce.58 Regulations on divorce were
subject to variance, depending on the social status of the wife and the terms of divorce.
But in the early days of the Republic, it was seen as a breach of loyalty for the wife to
copy her household keys, which likely contributed to infidelity and is seen as evidence of
promiscuous behavior.59
According to Cicero, the removal of the household keys is a finalization of
divorce proceedings. In speaking on the subject, he abstractly references foundational
Roman law. “In accordance with the Twelve Tables, he told his wife to take her
belongings, removed her keys, and drove her out.60” And though Latin scholars argue on
the punctuation of this sentence as a point of contention for the relinquishing of property
or materials, the confiscation of the keys is clearly understood to be a part of the original
law.61
The Status of the Roman Wife
The wife in the Roman household held any number of statuses, depending on her
background, wealth, or agreement of cohabitation at the time of marriage. The privileges
granted to the wife were dependent on her husband. In the instance where the wife’s
57
Plutarch Lives. 27-28.
58
Jane F. Gardner. Women in Roman Law and Society. (Bloomington and Indianapolis: Indiana
University Press, 1986), 83.
59
Apuleius, 9.20.
60
M Tullius Cicero The Fourteen Orations against Marcus Antonius. 2.28.69.
61
Gardner, 85.
32
status was under manus, she had very few rights, save for the ones afforded to her by the
husband.62 But her role was understood to be the mistress of the household, since most
husbands did not stay at home for employment, the lady of the house would oversee his
property and affairs.63 She would be granted the keys to the household as well as
oversight to the servants and slaves. To prevent pilfering, locks would be installed on
any number of cabinets or pantries and indeed, the large quantity of ring keys that survive
the archaeological record in addition to ancient commentary on the subject is implicit of
how widely these small locks were utilized.64
The master or mistress would undoubtedly wear or carry the rings that opened the
locks on his or her person. Since the lady ran the house in her husbands’ stead, it would
be understandable that the husband would grant her the possession of the household keys.
Even young girls were trained in this regard, being granted a set of keys for their own
small chest in the house’s storeroom.65 Social identity of the key-bearer was understood
in the ancient world as a symbol of status and power. If a lady wore the key-ring on her
hand outside of the house, it would identify her as the mistress to those with which she
interacts. The removal of the keys in divorce therefore held two principles; to remove
any access the wife had to the possessions of the house, and to strip her of her status as
mistress of the household.
62
Ibid., 11.
63
Gillian Clark. Women in Late Antiquity: Pagan and Christian Lifestyles. (Oxford: Oxford
University Press, 1994), 99-100.
64
Pliny, 33.26.
65
John Chrysostom On Virginity. 73.1 PG 48.586.
33
Criminality
Criminal behavior was severely punished in the ancient world. Breaches of
security through subterfuge, as played out by agents, appears to be commonplace and
cruelly punished. Ancient texts reveal that burglary and lockpicking occurred, though
most criminals rely on false keys to bypass security devices. Sallust’s Bellum
Jugurthinum relays an instance where a servant is manipulated into having false keys
made for a door to allow Jugurtha and his soldiers within at night to slay a king.66 These
false keys, or clavis adulterinae, appear to be physical copies made by a local
metalworker.
In the Metamorphoses, when Lamachus inserts his hand through a port to try and
unbolt the lock, the owner sneaks upon him and nails his hand to the door, trapping
him.67 According to the story, Lamachus’ men cut off his arm with a sword to make their
escape. Anyone caught in the act of a criminal offense could easily be punished under
Roman law, but an absence of evidence required the presence of witnesses. That
Lamachus was willing to sacrifice his arm to escape capture indicates the severity of
punishment awaiting him in the courts for breaking into a residence. A man may be put
to death if found guilty of theft by night, or if he steals during the day, armed with a
weapon.68
66
Sallust, 12.10.
67
Apuleius, 4.10.
68
M. Tullius Cicero Pro Marcus Tullius. 20.
34
The issue of theft is prominent in Ancient Rome and reveals the necessity for
public and private security. Latin texts delineate criminal theft, differentiating between
various types. Cutpurses and pickpockets are known as sacularii. Burglars of the kind
that would likely force entry past a lock by breaking it are referred to as effractores,
whereas those who do so without force are derectarii.69 The derectarii are likely those
who incorporated lockpicking into their criminal behavior. And though the penalties for
theft could be high, banishment may be interred if the crime is committed by an
individual of high social class. Mention is made of professional thieves, known as
expilatores, as being from upper class households.70 Upper class criminals faced
banishment rather than corporal punishment, due to their status. Laws were put into
place to stem the problem of merchants fencing stolen goods. Fencing was a form of
laundering, in which stolen money and goods are purchased by an intermediary and
resold in the market, allowing the thief a venue for his craft.71 Though locks could be
bypassed, they act as a hindrance and primary deterrent against criminal behavior.
Locking and Sealing
In Roman culture it is common to allow house slaves and servants possession of
keys for access to the lagers in the household. The tale of Myrmex in Metamorphoses is
indicative of this as is the instance of the doorman Euclides in Martial’s
69
O. F. Robinson. The Criminal Law of Ancient Rome. Baltimore: The Johns Hopkins
University Press, 1995), 27.
70
Ibid.
71
Ibid., 27-28.
35
Epigrammata.7273 The implementation of sealing objects may have been born in the
Roman period or just prior and is likely due to the misdemeanor of theft by slaves and
servants. “To think what life was in the days of old, and what innocence existed when
nothing was sealed! Whereas nowadays even articles of food and drink have to be
protected against theft by means of a ring: this is the progress achieved by our legions of
slaves…”74 Pliny laments the station of Roman possession as constantly under threat of
pilfering. Sealing was a practice that also protected the servant from unlawful
accusation. “For if all were well trained, there would be no need of seals, if servants and
masters were equally honest.75” The answer to the issue of theft and false accusation was
found in the locking and sealing of objects.
We can know that locking and sealing are used synonymously with one another to
prevent theft, as Pliny continues, “it is not enough to keep our keys under seal: while we
are fast asleep or on our death-beds, our rings are slipped off our fingers: and the more
important concerns of our life have begun to center round that tool, though when this
began is doubtful.”7677 Placing seals becomes common in the Roman period and is used
extensively in the later Byzantine era, but it is existent amongst the Greeks and early
72
Apuleius, 9.20.
73
Martial Epigrammata. 5.35.
74
Pliny, 33.26.
75
Clement, 3.11.
76
Pliny, 33.27.
77
Seneca De Ira. 2.25.3.
36
Egyptians as well.78 The locks of the tombs at Thebes have clay sealed over them,
marked by signet seals, which were found in situ.79 Objects and even locks could be
sealed with other mediums, such as wax or lead and even silver or gold.80 The true keyholders would therefore know by a broken seal if someone had tampered with the lock, or
even the sealed object. This would also prevent key-holders that were slaves from
accessing property and storage when the owner did not intend.
Likewise, Pliny the Elder relates a story of a woman that bypassed the lock on a
small casket to obtain the keys within to access the wine-cellar. Upon being caught, she
was locked away and starved to death by her own family.81 The casket that held the keys
was likely not locked, but sealed. Upon finding a broken seal, the owners knew someone
had tampered with the contents.
Keys and Eastern Texts
The oral law of Jewish custom reveals a contrasted structure for the wearing of
keys by women from Roman culture.82 “A woman shall not go out in public on the
Sabbath with a key which is upon the finger; if she does go out, it binds her [to make a
78
Aristophanes Thesmophoriazusae. 425.
79
J. Gardner Wilkinson. The Ancient Egyptians: Their Lives and Customs. Vol. II. (London:
John Murray Publishing, 1994), 364.
80
Vikan, 13.
81
Pliny, 14.89.
82
The Tosefta: Second Division, Moed. trans. Jacob Neusner (New York: Ktav Publishing
House Inc., 1981), Shabbath 4.11.
37
guilt offering.]”83 This passage reveals the social conservativeness of Jewish culture.
Women were not to flaunt their power and possession in public, at least not on the
Sabbath. The oral law reveals that women clearly held keys in their possession. No
direct evidence is known in regards to customary Jewish marriages to give comparison to
their Roman counterpart in key possession and divorce.
Keys in the ancient world are susceptible to breaking, just as modern keys. It
occurred often enough to be addressed in the Mishnah under the topic of cleanliness.
“If a knee-shaped key is broken at its joint it becomes insusceptible to uncleanness.
Rabbi Judah declares it susceptible since one may still open [the door] with the
inner portion. If a gamma-shaped key was broken off at its bend it becomes
insusceptible.”84
The ‘knee-shaped’ key is in reference to the tumbler keys that appear throughout
the Roman period. Such keys found in eastern cultures appear to have the same form
with some typological differences. The tumbler keys discovered at the Cave of Letters
were understood to be such keys, though they have a knob for a handle as opposed to the
loop which was common on Roman keys. Examples have also been labeled as ‘elbow
keys’, yet both labels are indicative of the common characteristic of a bend along the
body of the shank or handle at any point before the bit. Yigael Yadin remarks on the
observation of the Mishnah that an elbow key would still be considered susceptible to
83
Ibid.
84
The Mishnah. trans. Herbert Danby (London: Oxford University Press, 1933), Kelim 13.8.
38
uncleanness due to the function of the key, that even though it was broken along the
body, the bit could still be inserted into the lock.85
Yadin also mentions the gamma key as being characteristic of our modern keys
that rotate in the lock.86 The two primary key types found in the Roman period were then
likely used contemporaneously in the Near East. Though the Cave of Letters only
yielded tumbler keys, rotary keys were apparently in use as well. The Mishnah’s
composition was composed some time just before the turn of the millennium. This
implies that rotary keys were likely introduced to the Near East some time as early as the
1st century BCE.
85
Yigael Yadin. The Finds from the Bar-Kokhba Period in the Cave of Letters. (Jerusalem:
Israel Exploration Society, 1963), 94.
86
Ibid.
39
Figure 2: Hecate with Laconian Key
Image AN969959001
Courtesy of the Trustees of the British Museum
40
CHAPTER 4
LOCK AND KEY ANATOMY
Lock Anatomy
The nature of security in the ancient world relied on uniqueness and creativity to
construct effective locking mechanisms. Though every lock will vary in the precise
details of their construction, each form relied on general attributes which outlined their
function. These are separated into two categories: mount and casing. The casing of a
lock housed the internal components that interacted directly with the key in operation.
They were often exposed when installed on the interior of a doorway, but would be
obfuscated from view in their use with padlocks, cabinets, and chests. The mount
incorporates the elements of structure that connected the components to an object or
entryway.
The Mount
Lockplate and Keyway
The face of a lock is known as the lockplate. The lockplate prevents the
manipulation of the internal components by a foreign entity by limiting the access to a
specific key. The lockplate also serves the purpose of displaying aesthetic appeal. It is
one of the few parts, and often only part, visible after installation. The lockplate can
appear as a disconnected object, installed directly onto a structure. This occurs often in
situations where the door, chest, or cabinet is too dense for each component to be
41
connected. In situations where they need to be connected to the structure of the lock, like
small chests or padlocks, the lockplate often survives in context with the body of the
lock.
The identifying characteristic of a lockplate is often the keyway. In the tumbler
variety, the keyway is in the shape of an ‘L’, affiliated to the right, such as Figure 3. The
keyway of a rotary lock can vary, but is typically a vertical or horizontal slot, sized to the
structure of the key for which it corresponds. In archaeological analysis if a single
lockplate with a keyway survives, at least a single lock can be attributed to the unit if the
lockplate is in situ within an undisturbed context. In material analysis, the form of the
keyway is indicative of the lock taxonomy. For this reason, though most locks retained
from excavation and discovery are in a state of disrepair, a type can still be applied to the
object based on the lockplate and keyway. An escutcheon may appear on the lockplate to
cover the keyway from the elements.
Any material can be used to form the lockplate, though wooden locks typically do
not possess one, as is evidenced at Karanis. The utility of shaft keys requires a port
through which the arm and key are inserted to operate the lock. The surviving doors at
Karanis simply have a hewn recess in the door where the key was inserted directly.
Entryways without surviving doorways display qualities of the shaft key technology. 87
But lockplates are typically made from bronze, though the rest of the lock body may be
constructed of iron. Such an example was found within a Roman cave in Palestine, in
which a bronze lockplate and bronze parts were discovered amongst other materials,
87
Husselman, Plates 42 and 53.
42
including iron keys and iron objects associated with locks.88 In such cases, this is due to
a desire to portray quality and aesthetics on part of the metalworker. Bronze could be
polished to a lustrous sheen and was considered a precious metal. With a lower smelting
temperature, copper alloys such as bronze allowed for molds to be cast to create
ornamentation for bespoke objects.
More common in the late Byzantine period, but still found in the Roman period
are anthropomorphic or zoomorphic lockplates. Since such objects were often the only
parts visible, they were incorporated into the general appeal of the object for which they
were crafted, be it door, chest, cabinet, or padlock. Ornamentation appears to have been
added in the form of a lion emblem that was riveted onto a lockplate at Pergamon, in
addition to an unidentified zoomorphic latch.89 Socially, lockplates had the ability to
generate appeal as well as convey affluence in their form.
The Deadbolt
Though each component of the lock has an intrinsic role in the form and function
of each device, the deadbolt serves the primary function of securing. In the ancient Near
East, deadbolts were typically large wooden beams, hewn for the purpose of blockading
an entryway. The deadbolts found at Karanis were installed as a part of the lock casing
or door jamb of the frame.90 Such deadbolts could run horizontal or vertical, securing the
88
O.R. Sellers and D.C. Baramki. “A Roman-Byzantine Burial Cave in Northern Palestine,”
American Schools of Oriental Research 15-16 (1953): 18-26.
89
Machteld J. Mellink. “Archaeology in Anatolia,” American Journal of Archaeology 93, No. 1
(1989): 103-33.
90
Husselman, 41 and plates 43, 44, 45, 48, and 49.
43
deadbolt into the jamb of the frame or into the lentil and threshold, respectively.
Deadbolts secured small chests and cabinets by rendering the hasp of an adjoining lid
immovable.
The Homeric lock mentioned earlier is known to have used wooden deadbolts, as
such is preserved from various locations, including Egypt.91 Metal deadbolts may have
been used in the Hellenistic and Roman versions of this lock, because metal keys were
also being used by the Greeks at this time. The Roman tumbler locks relied on the
advantages of copper alloys and iron for the durability of their mechanisms. Such
deadbolts were constructed with various alterations for their function with locks and
keys. Because of the reliability of metal to withstand greater pressure, smaller deadbolts
and locking devices could be constructed.
The form of the deadbolt relays the form of the lock and oftentimes the key.
Grooved humps incised on wooden bolts or on metal keys accompanied with evidence of
attached rope or string is characteristic of the Homeric lock. Perforations within a
deadbolt can be characteristic of a tumbler lock. Wooden deadbolts had drilled holes to
receive the tumbler pins in the lock casing. The metal deadbolts of the Roman period are
unique and characterized by the preservation of their form and use with slide key tumbler
locks. (See Fig. 4)
The Hasp
91
James and Thorpe, 469.
44
Common on chests and likely cabinets, the hasp is an adjoining feature of smaller
locks that utilizes a separating lid. The hasp is joined to the lid of a container and has
free movement with the withdrawal of the deadbolt. Doors and entryways rely on the
frame or jambs as a secure receptacle for the deadbolt. The hasp performs the same
operation, but may come in the form of a U-shaped assembly that hooks the deadbolt
upon locking. Once the deadbolt was thrown with the key, the lid could be lifted, with
the hasp connected to its structure.
Examples of the hasp survive at the British Museum. (See Fig. 3) The hasp in
Figure 3 is still connected to the locking mechanism, but the organic material has
decayed. Excavations at Roman period Knossos unearthed elegant burial chests from the
1st-2nd Century CE which preserve this part.92 The hasp that survives on a chest that was
part of a burial utilizes the ‘U’ shape to secure the bolt. Retracting the deadbolt allowed
the hasp to be lifted as described. The presence of the hasp also leaves a distinguishing
characteristic on the lockplate in the form of a rectangular perforation, aligned with the
keyway. This is a diagnostic, which speaks to the construction of the hasp, as it was
necessary for the retainer to match up with the deadbolt.
The Casing
The casing of a lock houses the internal components with which the key interacts,
including the deadbolt in most instances. The casing would often be connected to the
interior portion of the door, not affixed through the door as modern locks are installed. It
92
Diana Wardle and K.A. Wardle. “Glimpses of Private Life: Roman Rock Cut Tombs of the
First and Second Centuries AD at Knossos,” in Knossos: Palace, City, State. eds. Gerald Cadogan, Eleni
Hatzaki, and Aonis Vasilikis, (London: British Schools at Athens, 2004): 477
45
would often be riveted or nailed into place, though evidence of wire has been found for
securing the casing to the body of an object.93
The casing of the earliest tumbler locks are comprised of wood, as are the internal
components. Homeric locks often do not have a casing, as such devices operate on few
parts, such as the deadbolt, rope, and key. The Homeric lock is installed with metal
staples, not an enclosed casing.94 Later locks had metal casings of either bronze or iron
to house the internal components.
Internal Components
Depending on the period, the actual components that comprise the internal
structure of the casing may vary and change. The tumbler technology utilizes tumbler
pins as a part of the casing to hold the deadbolt in position. The tumbler pins of the
Roman period are often made from iron and are pushed into position by a spring,
characterizing a change in technology. The spring applies force to the tumblers, keeping
them in a downward position, which is a significant variation from the first wooden
tumbler locks which rely on gravity to pull the tumblers down. Gravity-fed tumblers
likely met with resistance in their hand carved channels and may have not always fallen
into position into the deadbolt. The addition of the spring in later periods prevents this
from occurring. Keeping pressure on the tumblers also prevents them from being
jammed in their recess, so that they are always pushed back into position once the
93
Ibid.
94
James and Thorpe, 468-469.
46
deadbolt is slid back into the jamb and the key is removed. The various length and shape
of the tumblers corresponds with the tines for the key which operates them.
The spring itself is often a flat piece of iron, hammered thin with a slight re-curve,
affixed to the top interior of the casing. The re-curve of the iron creates the desired
spring-like action that applied pressure to the tumblers. (See Fig. 5) The first locks do
not utilize this technology and springs therefore appear at some point in the GrecoRoman period.
In the technological change that occurred in the Late Roman period, many of the
components remained the same with only two exceptions to the internal components.
The casing housed wards which obfuscated entry, taking the place of tumblers in its
action of security. While tumblers need to be manipulated to perform the action of
unlocking a mechanism, wards rely on obstruction in the rotation of a key to ensure the
appropriate key is used. Wards come in a variety of shapes and sizes to correspond with
the key and are often formed from sheeting and wire. A preserved warded lock in Paris
actually shows the rotary key in action, in addition to the wards that barred keys from
access.95
The additional component found in rotary lock casings was a mounting recess or
mounting protrusion for the muzzle of rotary keys. Such mountings are not common
with tumbler locks and are used for the function of stability. When the proper key is used
in a rotary lock, it will rest comfortably in, or on, the internal mount and will allow for
proper rotation.
47
Key Anatomy
The physiology of the key can be divided into three general sections: the handle,
the shank, and the bit. All three sections play an integral role in the function of the key
and provide information on their use after discard. The handle relays information on how
the key was carried by the user and also how it was held in operation. Secondary features
are sometimes incorporated into the handle that retain little functional use, but rather
display general aesthetics or indications of affluence as a bespoke object.96
The shank and the bit relay the direct functional use of the key as their design
corresponds with the form and structure of the lock. The bit is often the primary
indicator of lock and key taxonomy, and when found outside of the context of the lock,
can give provisional information for a reconstruction of the lock’s internal components.
The Handle
Hoop
Just as today, Roman keys needed to be attached to a retainer to prevent the owner
from losing the object. The hoop is a common characteristic that appears in all keys in
one form or another and its use is in securing the key to the body. It is likely that keys
________________________
95
Singer, 429-30.
96
For those conducting personal or professional research on the subject of locks and keys I
recommend that particular attention be paid to the bit of the key. While keys vary significantly in size and
shape, the function of the key is held primarily in the bit. The features of the bit provide insight into a keys
use, and though keys are more often found alone than in situ with the locks for which they are used a
deconstruction of the physical properties of the bit will reveal the taxonomy of the lock. Just the same, the
physical form of a lock and its construction will at minimum pronounce the taxonomy of the key, if not the
exact shape and form for reconstruction. The excavators of Knossos did an outstanding job of this by
reconstructing a slide key based on the surviving lock of a small Roman chest. See: Wardle, 479.
48
were worn on leather thongs draped around the neck, or on metal rings worn at the waist.
Typical of the Roman period and beyond, keys were also worn on the finger to safeguard
the object from loss and to also display wealth and affluence, as was discussed in the 3rd
chapter.
The hoop displays attributes of decoration or design, particularly in instances
where the key is crafted as a mold from bronze. With the early iron keys of the Roman
period, the hoop is often inlaid into the body of the handle. This is due in part to the
construction of the key and the material, as iron could not be cast into a mold and
therefore had to be heated and shaped to form the hoop. This would have likely been
accomplished on some form of an ancient mandrel, just as modern rings are sized and
shaped. Keys made from bronze tend to have more distinguished hoops, set apart from
the body of the handle and often being attached by a hinge, a common practice from the
early Byzantine period until the medieval age.
Flange or Bezel
A protuberance often appears on the butt of key hoops that, except in the
Byzantine period, served little purpose other than decoration. These extensions from the
end of the hoop come in two forms, the flange or bezel. The flange is simply extra
material which may have been a part of the decorative features of the hoop, while a bezel
is an indicative feature of signet rings. The bezel has a slightly raised platform, often in
the shape of a cone or orb. On signet rings, the bezel was incised with inscriptions and
49
designs, to be used as an impression for a sealing medium.97 The addition of bezels to
keys appears to be an indicator of form change in the late Roman to early Byzantine
period, as keys made in the Byzantine period are more inclined to have a bezel than a
flange. This may be in part due to a socio-cultural change, as the sealing of locks and
objects with wax and other mediums becomes more commonplace in later periods,
though mention of the topic is made from the middle period of the Roman Empire.
Keys from both the Roman period and the Byzantine period do not always display
a flange or bezel. In many instances the key itself simply retains a hoop attached to the
shank, with little to no decoration. And though the bezel appears more frequently in
Byzantium, not every key necessarily retains a bezel. The primary function of the bezel
holds a secondary countermeasure with the addition of sealing, as previously discussed.
Grip and Neck
Generally found just below the hoop of the key, the grip and the neck are defined
as the area of the handle that is held during operation. The namesake of the grip implies
its function as the point of the key that is held, and the point of termination where the grip
meets the bit of the key is the neck. The features of the grip are generally ergonomic, but
can be decorative. Bespoke handles are often created in the shape of anthropomorphic
and zoomorphic grips. (See Chapter 8 for examples)
The neck is found at the point of termination for the grip of the key. It marks the
position for which the handle meets the shank and bit. The neck can be clearly known in
situations where the handle is cast and joined to the bit, much in the same way that a
97
Vikan, 13.
50
handle is affixed to a blade. Such artifacts serve as examples of mass production, in
which a metalworker forged multiples of the same handle and attached them to bits to
function in various locks. This variation can be identified by a recess for the shank that is
found at the end of the grip.
When a key is used in a lock, whether rotary or tumbler, pressure is applied to the
key from the handle to the internal components. The point of tension is applied to the
neck and is often the point at which a key fractures and breaks.98 We know from literary
sources that ancient keys did have a tendency to break, even the hefty tumbler keys of the
early 1-3rd Century. If enough torque is applied to the grip, the neck of the key will
break, just as modern keys often do.
The Shank
Flat or Barrel
The form of the body that is directly attached to the bit can be described in two
ways; as either a flat shank or as a barrel shank. The flat shank form is square or
rectangular in shape, but does not conform to a specific size or length. The shanks of all
keys vary, depending on the depth to which the key must be inserted into a lockplate to
operate it. Barrel shanks have a rounded form and their shape follows the function of
rotary keys. Generally, if the body of the key is a barrel, it is most likely a key used for a
rotary lock. Conversely, if the body of the key is a flat shank, the key may have been
98
This is evidenced by the large number of keys that are missing either the handle or the bit in
a variety of collections. The bronze key bits of the Byzantine period are often missing their respective
handles, due to a fracturing that occurs at the neck. Dozens of such examples exist at the Menil Collection
in Houston, Texas.
51
used in a tumbler lock, though in some instances this does not appear to be the case. The
key in Figure 6 appears with a flat and rectangular shank, but clearly has perforated holes
for wards of wire. In utility, such a key would be more difficult to operate, lacking a
rounded barrel body for proper rotation.
The Bit
As mentioned before, the bit of the key indirectly relays the form and function of
the lock for which it was made. The bit is typically found at the end of the key and is the
point of contact for the key and the internal components of the lock. The shape and
design of the bit corresponds with the countermeasures installed in the lock that prevent
unwanted entry. Depending on the form of the lock, the bit may retain a variety of
attributes which differentiate the key from others and define its utility.
Tines
Indicative of the keys that operated tumbler locks are tines. Tines appear much
like teeth on a key and come in a variety of shapes and sizes. The length of the tines, as
well as their number and distance from one another correspond with the tumbler pins that
would have been installed into the lock. For this reason, tines play a significant role in
both knowing and recreating the internal construction of the lock. Tines are allocated
specifically to their use in tumbler locks.
Perforations, Grooves, and Channels
Though not specific to the function of any singular lock form, perforations,
grooves, and channels appear most often on rotary keys. This is due to their purpose to
52
bypass the wards installed as part of the internal components of the lock. On rare
occurrences, warding of tumbler locks may have occurred, as is evidenced by the keys
whose bits incorporate both tines and perforations. (See Fig. 17)
The primary function of perforations is in its use with the rotary lock, but they can
take on decoration to grant the key aesthetic appeal. Cruciform perforations become
common after Constantine and well into the later Byzantine period, as the advent of
Christianity began to influence the Eastern Roman Empire. Perforations as inscriptions
seem to be rare, but are at least known to have occurred in one instance. Figure 7 is a
rotary key with a perforation in Greek, spelling out a name, possibly of the owner.
Grooves appear to be similar to tines, but differ slightly in their form and use.
Elongated grooves that do not function as tumbler tines may be indicative of a variety of
the rotary lock. This form may easily be confused for tumbler technology. The groove
diagnostic will be addressed more thoroughly in chapter 6 on the rotary form.
Though similar to perforations, channels appear as separations on the bit along the
shank. The form gives the key the illusion of having more than one bit along the body of
the key, but this is not the case. The channel serves the same purpose as perforations,
bypassing a larger ward within the lock. The obfuscation of a lock by a larger ward
would prevent most keys from being operated in the lock that did not have a large
channel. This may appear as an added layer of security, but it is simply another ward and
a prominent characteristic of the key.
Platform
The foundation of the bit, upon which tines and grooves are etched or incised, is
53
designated as the platform. The platform serves the purpose of establishing the
dimensions of the key bit. The height and width of a platform will never be larger than
the keyway will allow.
54
Fig. 3: Early Roman Lockplate.
Image AN103710001
Courtesy of the Trustees of the British Museum. London.
Fig. 4: Roman Deadbolt.
Image AN921576001
Courtesy of the Trustees of the British Museum. London.
55
Fig. 5: Roman Tumbler Lock (Reconstruction)
Courtesy of the Deutsches Museum, Muenchen.
Fig. 6: Rotary Key
Image AN908091001
Courtesy of the Trustees of the British Museum.
56
Fig. 7: Rotary Key with Greek Inscription
Image AN921664001
Courtesy of the Trustees of the British Museum.
57
CHAPTER 5
THE TUMBLER FORM
“I decided that the best thing to do was to sneak away just before daybreak and
start travelling, even with shaky steps. I picked up my little bag, pushed the key
up into the lock, and tried to slide back the bolts. But those good and faithful
doors, which had unlocked of their own accord during the night, now opened only
with enormous difficulty and a long effort and many insertions of the key.99”
Tumbler locks characterize the most commonly utilized lock in the Roman period
and form the foundation for changes in lock technology beyond this period. But the
Romans were not the first to advent this technology. The passage of tumbler technology
through Ancient Mesopotamia and Egypt into Western Europe is not remarked upon by
ancient historians. Despite that it is known that the oldest surviving mechanisms come
from those regions, such technology is attributed to a western origin, as Pliny’s
attribution of the invention of such devices is given to a Hellene.100
Tumbler Attributes
In identifying Tumbler technology, the most distinguishing feature to note is that
of the tines found on the bit of the key (See Fig. 10). As discussed in the anatomy of
locks and keys, each of these tines corresponds with a complimentary tumbler pin in the
locking mechanism. While the Mesopotamian lock utilized a known wooden technology
that relied primarily on gravity to keep the tumbler pins down in the locked position, the
99
Apuleius, 1.14.
58
Romans often added thin sheets of metal as springs. These springs applied resistance to
the pins, to ensure that they remained down in the locked position until pressure was
applied to them from the tines of the key. The shape of the tines correspond with the
recess for the tumbler pins, ensuring the security of the lock if the key is not cast several
times in a mold and is uniquely made.
Tumbler locks came in either a one-handed or two-handed operation format. The
one-handed operation allowed the user to insert the key directly into the lock, lifting or
sliding the deadbolt out of position for entry. Two-handed tumbler locks required more
dexterity, as the key had to be inserted into a position to lift the tumblers, using the other
hand to throw the bolt. Indications of the two-hand tumbler lock reveal that oftentimes a
port-hole was cut into the door to allow passage of the hand and key to the locking
mechanism on the inside. The period of the Roman Empire sees a progression of tumbler
technology to a one handed operation with a lock-plate situated on the exterior of the
door, as the lockplates in the archives of the Deutsches Museum in Munich date to the
1st-2nd Century CE. (See Fig. 16) If only the key remains, it may be difficult for an
analyst to ascertain the format for which it was used, but some indicators shed light on
this matter.
One-Handed Tumbler Lock
These locks generally have an L-shaped recess on the front of the lock-plate to
receive the key. The key would be inserted into the recess and, upon displacing the
tumbler pins, could drag the deadbolt down or over to remove the deadbolt from the
________________________
100
Pliny, 7.198.
59
frame, thus unlocking it. In much the same fashion, the deadbolt would be locked back
into place with the key. Keys of this nature have been referred to as slide-keys, and
though commonly used throughout the Roman Empire, are unique in form and utility.
But shaft keys are often used for such lock types and still required that the key be inserted
through a port in the door. The reconstructed tumbler lock and key in Figure 5 is a
standard form for one-handed operation.101
Two-Handed Tumbler Lock
The two-handed tumbler lock provides some difficulty in identification. Both
locks, while following the same principles of operation vary slightly in their mechanical
assembly. The access for the key within a one-handed tumbler lock rests directly in the
deadbolt itself, while the two-handed tumbler lock removes the access from the deadbolt
by means of insertion within the tumblers themselves. If the tumblers survive it would be
possible to identify it as part of a two-handed tumbler lock, as the point of access for the
key is carved into the body of the tumblers. A drawing of such an example can be found
in Life in Biblical Israel.102 With the key inserted into the body of the tumbler pins, the
second hand would be able to pull a latch attached to the deadbolt, to withdraw it from
the door jamb.
To carve out a channel into the deadbolt that would accept the passage of the key
and tines would require a great expenditure of time and resources. Considering that such
101
A written diagram of the tumbler format can also be found in the Guide to the Antiquities of
Roman Britain, by J.W. Brailsford, page 76.
102
Philip King and Lawrence Stager. Life in Biblical Israel. (London: John Knox Press, 2001),
33.
60
a recess would require a significant amount of room for the key and space for the
metalworker to form the passage, it might also be assumed that this affected the stability
and structure of the deadbolt itself. But the choice of a one-handed lock over a twohanded lock is relative to the metalworker and the patron. The stability of the lock in
comparison to the time and resources needed to create the object were likely determined
at the time it was commissioned by the patron.
The tumbler lock was the standard throughout the Roman period, though it varied
in form between the one-hand and two-hand tumbler. Such locks were constructed of
iron, with external plates and parts made of bronze. The reason for this mode is likely
due to cost and appearance. Bronze was a precious metal and could also be highly
polished. Any part exposed would grant a far greater aesthetic appeal than cold iron,
which varied in appearance and did not garner the same golden appeal in the visage
which bronze offers. This is evidenced where both the external and internal parts of the
lock survive.103104
Key Forms for Tumbler Locks
Keys for tumbler locks will often display any of the attributes listed in the handle
and shank found in the anatomy section of this typology, except for the hinge. The innate
function of keys for tumbler locks is to apply pressure to spring-loaded, or gravity fed,
tumbler pins and the swivel action of a hinge constrict this ability. For the bit,
perforations, grooves, and channels are non-existent as the technology relies on tines to
103
Wardle, 476-79.
104
Sellers and Baramki, 7-55.
61
interact with the internal components of the lock. To allow for appropriate pressure
against the tumblers, the shank of the key is often a solid form of metal. But not all keys
made for tumbler locks were composed of a solid state of metal.
Keys recovered through various periods share similar construction facets with
swords and knives, possessing a tapered shaft that is fit into a handle stock. An example
survives from Troy, in which Heinrich Schliemann unearthed the bit of a copper-alloy
key with a tapered shaft for a wooden handle.105 Copper-alloy handle stocks, of bronze
composition appear to be used quite frequently with hammered iron. The handles of keys
found in collections which are missing the bit component often have iron remnants or
bear the features of handle-fitting in their characteristics, such as Figure 8. Handles are
easily mass-produced and provide a metalworker with a generic stock to fit onto the keys
constructed. Theoretically, handles are an indication of mass production, thereby
implying that a key made from a solid state and not joined to a stock are more inclined to
be a bespoke artifact.
The Laconian Form
Indications of tumbler technology contemporaneous with the Roman period and
the preceding Hellenistic survive in the musings of ancient playwrights and orators. In
several ancient texts where keys are mentioned, a Laconian origin is given. In Plautus’
Mostellaria, a door is locked specifically by a key labeled as Laconian in form.106 This
105
Heinrich Schliemann. Troy and Its Remains: A Narrative of Researches and Discoveries
Made on the Site of Ilium and in the Trojan Plain. (London: John Murray Publishing, 1875), 333.
106
Titus Maccius Plautus Mostellaria. 2.1.57.
62
implied a certain identifying attribute about the key that grants an origin from Sparta.
Thankfully, at least one of these attributes is known from external texts. In
Thesmophoriazusae, one of the women speaking before the Thesmophoria on the change
in social status of women remarks on the possession of keys in the household. “Our
husbands now carry little Spartan keys on their persons, made with three teeth and full of
malice and spite.107108” The Laconian keys are distinguished as having three ‘teeth’.
This description follows the form of even the Egyptian keys at Karanis, which also have
three separate tines. Roman keys differ in this regard, as the common tumbler key has
significantly more tines than the preceding Laconian key, undoubtedly to increase
security.
But the Laconian key remains iconic, even into the turn of the Millennium. A
stucco-painted wooden tablet from Egypt which survives from sometime between the
Ptolemaic and Roman Periods reveals a crouching figure wielding a three-tined key. (See
Fig. 9) The form of the key is typical of a slide key, though its relative size to the person
is comparable to the larger types of utilitarian keys found in the Roman period. A
Laconian style key survives from Heinrich Schliemann’s excavation of Troy, composed
of iron, bearing three tines.109 A similar key to the one discovered at Troy also survives
107
Aristophanes, 420.
108
Further ascriptions to Laconian keys do not survive in historical writings of the Greek and
Roman periods. A secondary reference through Eustathius, the 12 th Century commentator on Homer, may
indicate a similar understanding of Hellenic key origins in Sparta, as well as with Suid from the third
century BCE. See Eustathius, Commentary on the Odyssey, p. 1603, 51. See also Müller, The Archeology
of the Dorians, p. 25.
109
Heinrich Schliemann. Ilios: City and Country of the Trojans. (New York: Harper Brothers,
1881), 622.
63
at the Egyptian tombs in Thebes, bearing three tines.110 The Laconian form likely fell out
of fashion and use amongst the Romans, though may have still been utilized. A Laconian
key with three tines is displayed in The Finds of Roman Britain, amongst the materials
recovered from Richborough.111 The Roman period maintains a large swathe of tine
variations, all of which vary in size, shape, and number of tines. This is indicative of a
desired change in the tumbler form, as the Laconian form is likely not complex enough to
bar entry from criminal behavior.
The Slide-Key Form
No key form appears in higher volume throughout Late Antiquity than the slidekey form. The standard of locks in Roman culture, the slide-key could very well be an
improvement from the three-tined Laconian form, and allowed metalworkers the
potential for a greater variety in their locking mechanisms. The slide-key is also
versatile, as the technology can be used in tumbler locks installed on doors, chests, and
cabinets.
Though no exact chronology has been established for locks and keys, it is likely
that the slide-key became more prevalent at some point during the Roman Republic as
such keys survive in antiquity from before the 1st Century BCE and persist into the
middle Byzantine period of the Eastern Roman Empire. The Menil Collection has a
variety of keys dating to Byzantium which exhibit the slide-key function and have been
110
Wilkinson, 16.
111
Guy de la Bédoyère,. The Finds of Roman Britain. (London:B.T. Batsford Ltd., 1989), 114.
64
remarked upon by Dr. Gary Vikan.112113114 Such keys hold the same diagnostics in their
attributes and composition, making it difficult to date a key outside of a known
archaeological context.
Following the anatomy in Chapter 2, the slide-key can display most of the
attributes found in the handle and the shank, except for the swivel. The body of the key
terminates to a distinct “L” shape at the bit, which corresponds to its insertion and use
within the lockplate. (See Figures 10, 11, 12, and 13) The four lockplates recovered with
the Roman Tal-y-Llyn hoard have the diagnostic “L” indicative of slide-key use, two of
which also have the retaining passage for a hasp, indicating their use on a chest.115 The
tines vary in shape, size, and placement on the bit to match the perforations of the
deadbolt (See Fig. 14) They range in number, typically from four to nine and can be
found in multiple publications.116117
The Lift Form
Sometimes designated also as a latchlifter key, the lift form of a key is currently
the only surviving form with a diagnostic of its use for two-handed tumbler lock
112
Gary Vikan. “Unpublished Chapter on Keys and Locks from a Book on the Byzantine Small
Finds in the Menil Foundation Collection.” (Houston:Menil Collection Archives, 1994).
113
Vikan and Nesbitt, 1-7.
114
Made of Iron. September-December 1966. University of St. Thomas Art Department.
(Houston:The Menil Collection, 1966), 71.
115
Mansel G. Spratling. “The Date of the Tal-y-Lyn Hoard,” Antiquity 45 (1966): 229.
116
Alan McWhirr. Roman Crafts and Industries. (Aylesbury: Shire Publications, Ltd., 1982),
53.
117
See also: Made of Iron and Bédoyère, 114.
65
operation.118 The bit of the key is inverted, with the tines pointing towards the shank and
handle, instead of being aligned in an “L” shape common for slide-keys.119 (See Fig. 15)
Such a key is inserted into the port of a door and passed over the body of the locking
mechanism, inserted from below to lift the deadbolt out of a locked position as opposed
to sliding. This could potentially be for deadbolts which secured entryways vertically
into the lentil or the threshold, as opposed to the common horizontal installation into the
doorjamb. The inverted bit is the only diagnostic which designates an artifact as a liftkey, as the other features of the key can be common with the Laconian and slide-key
forms. Several examples of the lift key form survive from the Roman iron hoard at
Silchester, though each bear only two tines.120 A single lift key was recovered from Firka
which bears resemblance to Fig. 15.121 The few hellenistic keys found in this research
appear to follow the lift form, but their exact provenance and time period is unknown.122
Ring-Keys for the Tumbler Form
Cabinets and chests had smaller locking mechanisms for their security.
Corresponding with their size, ring-keys are likely used for such instances and appear in
significant number. Ring-keys made for tumbler locks follow the attributes of the
standard key and should be designated as such with the addition of ‘ring-key’ to any
118
Timothy Darvill, ed. The Concise Oxford Dictionary of Archaeology, 2nd ed. (Oxford:
Oxford University Press, 2008), s.v. “Latchlifter”.
119
See also: Made of Iron, 71.
120
McWhirr, 49, 53.
121
Lawrence P. Kirwan. The Oxford University Excavations at Firka. (Oxford and London:
Oxford University Press, 1939), 32.
122
See also: Vincent J.M. Eras, Locks and Keys throughout the Ages, p. 36, Fig. 31
66
analysis for an artifact worn on the finger. Evidence of surviving keys appears to only be
of the slide-key form, as examples of the Laconian and lift-key types were not found in
this research. Lift keys are used on larger locks and deadbolts and are therefore, by their
nature, too cumbersome for use in a small chest or cabinet. Though no Laconian form
has been found in the ring-key variety, it does not mean they do not exist, only that such a
form cannot be ascribed to this typology at this time. The operation of a ring-key for the
tumbler form follows the same function of its larger counterpart. The only difference lies
in its compact versatility to be worn on the finger and be used on smaller locks, in
addition to the social implications they held.
67
Fig. 8: Roman Key Handle
Access. # 06.176.12
Copper-Alloy, 1st-7th Century
Courtesy of the Metropolitan Museum of Art
Fig. 9: Crouching Figure Holding a Laconian Form Key
Access. # 13.182.45
Ptolemaic or Roman Period, 4th Century BCE – 3rd Century CE
Courtesy of the Metropolitan Museum of Art, New York.
68
Fig. 10: Late Roman Slide Key
Access. # CA 64086
Hickey-Robertson, Photographer
Courtesy of the Menil Collection, Houston Texas
Fig. 11: Roman Slide Key
Access. # CA 64098
Hickey-Robertson, Photographer
Courtesy of the Menil Collection, Houston Texas
69
Fig. 12: Early Byzantine Ring Key for Tumbler Lock
Access. # X 490.549
Hickey-Robertson, Photographer
Courtesy of the Menil Collection, Houston Texas
Fig. 13: Roman Slide Key
Courtesy of the Trustees of the British Museum
70
Fig. 14: Roman Slide Key with Deadbolt
Courtesy of the Trustees of the British Museum
Fig. 15: Roman Lift Key
Access. # CA 64087
Hickey-Robertson, Photographer
Courtesy of the Menil Collection, Houston Texas
71
Fig. 16: Roman Lockplates and Reconstruction. (Bottom Right)
Courtesy of the Deutsches Museum, München
72
CHAPTER 6
THE ROTARY FORM
Rotary Locks
While most locks in the Roman period evolved, or were borrowed from ancient
forms, the rotary lock is endemic to Roman technological advancement. Rotary locks do
not have a precise origin before their appearance within the archaeological record at this
time, and there appear to be only limited samples of rotary locks or keys before this
period based on the few Hellenistic examples found within the archaeological record.123
An Egyptian sarcophagus key from 300 BCE has diagnostic features indicative of the
rotary form, and may be one of the earliest known rotary keys.124
This lock form is characterized by its construction and operation. While tumbler
locks relied on tumblers and tines for lock and key manipulation, rotary locks function on
a similar principle of modern locks. The key is inserted into the lockplate, guided by
either a mounting pin or recess. Once the key is correctly fitted into the body, a simple
rotation of the key with the hand would force the bit of the key to contact the deadbolt,
pushing it out of the locked position as the key turns. This structure varies significantly
from the tumbler lock, as it does not rely on tumbler pins or tines to bypass the security
features, but incorporates the technology of warding.
123
Schliemann, Ilios., 620.
124
Eras, 39.
73
A rotary key may be labeled as a ‘warded key’ in a collection or analysis, but this
is a misnomer, as the key itself is not warded. The ward is found within the body of the
lock and the name of the key should follow its operation to avoid confusion. In addition,
tumbler locks may have utilized wards at some point, as some keys exhibit both
perforations and tines. The same key example as Figure 17 can be found in Chiavi e
Serrature.125 Use-wear analysis will have to determine points of impact to see if the key
was rotated within a lock or struck upwards to determine if this is indeed the case.
Wards can be defined as any object or structure invested into a lock to prevent the
access of any key other than the one intended.126 These simple additions prevent most
keys from being inserted, other than the one prescribed. Tumblers should not be
confused with wards. Though they serve a similar purpose in security, tumblers are
manipulated as a part of the object’s function, whereas wards are static and do not move
and ensure that the only the correct key will operate the lock. Wards come in a variety of
shapes and sizes. No standard seems to exist amongst ancient metalworkers for the
construction of wards, but certain variances appear to be common and recurring
throughout history. The two major forms of warding a lock can appear on either the face
of the lockplate, or within mechanical interior of the lock.
Warded Lockplate
The first wards that appear for rotary locks formed the face of the lockplate as the
primary security feature for ancient keys. Tumbler locks and the slide keys made for
125
Giorgio Lise. Chiavi e Serrature. (Milan: BA-MA Editrice, 1987), 5.
126
Eras, 1-12
74
them relied on the “L” shape keyhole for the lockplate. This was utilitarian, not based on
security. Rotary locks in the Roman period may have come in a variety of shapes and
sizes and served the function of primary security measure. The face of the lockplate
would be designed specifically for the shape of the key.
Based on the deconstruction of rotary locks, anything could be inserted into the
lock that might be able to throw the deadbolt, since the deadbolt is operated by simple
rotation and not tumbler retraction. The metalworker’s response to this was to shape the
keyway of the lockplate so that only the correct key could fit through it. This prevented
someone with a different key from operating the rotation feature of the deadbolt, but
would not fully prevent it. A burglar could easily bypass such a ward by using or
creating a minimalistic tool to insert into the lock and throw the deadbolt. The design of
the warded lockplate exhibits a distinct change in technology, but exhibits an inherit
design flaw in its security. The aforementioned problem is characteristic of any type of
rotary lock in the ancient world, and likely persisted to the modern age, until lock
technology finally changed in the 19th Century. It is plausible that wards began to appear
on the interior of rotary locks from the late Roman period into the Byzantine era to
address and minimalize unwanted infiltration.
Warded Locks
The addition of interior wards would certainly increase the general security of a
rotary lock, but would not completely secure it. A thief seeking means of bypassing such
a lock would face more obstacles in approaching internal wards, but since the function of
75
interior wards does not change the operation of the lock, the security of the device is still
susceptible to manipulation.
Such wards may have been coupled with a warded lockplate to increase security,
but surviving examples of rotary locks generally maintain a simple opening to allow the
passage of the key onto its recess for rotation. Interior wards varied in size and shape.
They could be composed of plating and wires that correspond with the key. From this
research, currently only one surviving warded lock has been discovered that displays the
inner workings of the rotary form and is on display at the Louvre.127 The wards
themselves were constructed as part of the interior of the lock and the relevant key would
be shaped for its purpose. The wards would pass through the bit and body of the key
when it was inserted and rotated.
Rotary Keys
Though there are few warded locks to provide a proper sampling of the data for
analysis, rotary keys survive in relative abundance and can be used to grant provisional
information on the rotary form. Just as the tines on a slide key can speak to the shape and
construction of the lock for which it was used, rotary keys do just the same. These keys
vary significantly in size and form, but carry distinct perforations on the bit that define
their use. If the bit survives on the key, the wards can be realized and even reconstructed.
Wards that pass through the perforations of the key likely span the rotation of the lock.
Wards would be effective for preventing the wrong key from passing through the
127
This can be found in A History of Technology, Vol. II pages 429-30 By Singer, Holmyard
and Hall.
76
mechanism, but it was likely that some keys may have worked on locks other than the
one for which they were intended.
The earliest rotary keys appear simplistic in nature and design, with simple bits
and few perforations. An anthropomorphic rotary key from Ilios has one large
perforation within the bit.128 Ornate rotary keys appear in greater number in the
Byzantine period, with increased complication in bit structure and shape of perforation.
Cruciform perforations are fairly common in the Christianized Eastern Roman Empire.
(See Fig. 18)
Grooved Rotary Key
Keys that do not possess tines may be characterized as rotary, but differentiating
between a rotary key for a warded lockplate versus a warded lock must be defined.
Firstly, it must be understood that a key can display both attributes, as the lock can
contain interior wards as well as a warded lockplate. But they can also have one or the
other. A key for such a lockplate will maintain a distinguished bit, but without tines or
perforations. In such cases where an example survives, the bit is irregular, seeming to be
a simple platform but with grooves replacing the tines, such as Figure 19. For such a key,
grooves and incisions mark the shape of the bit to allow its passage through the lockplate
and into the recess beyond. Several keys recovered from Pompeii and Herculaneum are
incised and grooved for use in the warded lockplate design.129
128
Schliemann, Ilios., 621.
129
See Vincent J.M. Eras, Locks and Keys throughout the Ages, p. 35, Fig. 29.
77
Perforated Rotary Key
Due to the unreliability of the lockplate as the only countermeasure to security
infractions, additional warding would be necessary. Once interior wards were added to
locking mechanisms, the form of the key evolved significantly. Key bits for rotary locks
begin to take a standardized rectangular or square shape during the late Roman period
into early Byzantium. The bit would often have a rectangular or square protrusion at the
end of the shank, with rounded edges. (See Figs. 18 and 20) It would contain
perforations for the corresponding wards of the lock, taking various shapes and
placements on the bit. Such wards may have been at the behest of the patron, as bespoke
objects are not unheard of, to add particular shape or design to the commissioned work.
Though the shapes and placement of the perforations vary from key to key, an
examination of a multitude of rotary keys reveal that some general principles appear to
have been followed. Only holes and designs were carved that likely would have flowed
naturally through the lock and did not go beyond the boundaries of a metalworkers
capabilities. Metalworkers in the Roman period were working with sheet metals, rods,
wires, and ingots. The limitations of the metalworker spanned the availability of tools
they possessed and the shapes they could form from their supply. For this reason, most
perforations appear to have simplistic gaps and round holes, for the installation of iron
wire and thinned iron sheeting as wards, as in Figure 4. Iron sheeting and even bundled
wire could be installed as a cruciform ward. Most wards and lock interiors were
comprised of iron and due to their susceptibility to ferric oxide, do not often survive.
78
Approaching the Byzantine period, key shanks begin to have increasingly
shortened body forms and the loops are then formed as an external component, attached
as a hinge as opposed to integration with the body of the key. (See Fig. 21) No textual
evidence is known for this typological change, but it is likely due to the fragile nature of
the rotary key itself, addressing a common issue of broken shanks and handles.
Slide keys may have been less prone to breaking off in a lock because of their
dense and rigid form. An examination of the keys from the Menil Collection in Houston
has revealed that the shanks and bodies of slide keys are greater in width than those of the
rotary variety, which commonly are barrel shaped and have a narrow circumference. The
torsion applied to a rotary key would occur at the neck of the shank where it meets the
loop. Constant use and aggressive force on a key within a stubborn lock would cause it
to break at that point. In many cases where keys are recovered, the loop is missing from
the shank, or only the bit survives. Such examples appear to be more prevalent in
archaeological salvage from the early Byzantine period onward; as such keys became the
standard for construction and use.
Rotary Ring Key
It is not uncommon for rotary keys to come in the ring key form. (See Figs. 22
and 23) The preference of technology in choosing either a tumbler or rotary lock as a
ring key was dependent on either the consumer or the metalworker. Such ring keys likely
served the same function as the keys for the tumbler counterpart in its use on chests or
cabinets. The bits vary in size and shape, but the conceptual format of rotary locks allow
for smaller lock installation. Though rotary keys become standard amongst ring keys in
79
later periods, the transitional phase that takes place in the mid to late Roman periods
displays examples of both tumbler and rotary ring keys. The anatomy of the ring key
possesses the same attributes of its contemporary rotary key type. The bit contains
perforated markings to function with the installments of wards in the smaller versions of
locks. Ring keys typically lack an extended shank, oftentimes wielding either a short
shank or swivel attachment connected to a ring to be worn on the finger.
80
Fig. 17: Roman Slide Key/Rotary Key
Image: AN257751001
Photo Courtesy of the Trustees of the British Museum
Fig. 18: Early Byzantine Rotary Key
Access. # X 490.539
Photographer: Ursula Pariser
Courtesy of the Menil Collection, Houston, Texas
81
Fig. 19: Early Byzantine Rotary Key with Grooved Bit
Access. # X 490.562
Photographer: Ursula Pariser
Courtesy of the Menil Collection, Houston, Texas
Fig. 20: Middle Byzantine Rotary Key
Access. # X 490.541
Photographer: Ursula Pariser
Courtesy of the Menil Collection, Houston, Texas
82
Fig. 21: Rotary Key
Image AN921660001
Courtesy of the Trustees of the British Museum.
Fig. 22: Rotary Ring Key
Image AN908082001
Courtesy of the Trustees of the British Museum.
83
Fig. 23: Roman Ring Key
Acces. # CA 64097
Photographer: Hickey-Robertson
Courtesy of the Menil Collection, Houston, Texas
84
CHAPTER 7
THE PADLOCK
For centuries before the Roman period, signet rings and cylinder seals were used
to protect objects and documents during transit. Though such objects are guarded from
theft and alteration on part of the courier, they are not secured from external forces. But
the invention of the padlock during the Roman period provided the means to restrict
access and introduced the concept of portable security to the ancient world. Brought to
the Roman frontier at some point around the turn of the Millenium, the padlock appears
to be a purely Roman invention for Western Europe, though such devices do begin to turn
up in isolated instances not long after the 1st Century CE in East Asia.130 The function of
the padlock components can differ from the typical Roman lock. A diagram of the
padlock components and function can be seen in A Guide to Roman Britain, as well as
Roman Crafts and Industries.131132 The padlocks found for this research are found
predominantly in Great Britain. The prevalence of padlocks in Britain may be due to the
necessity of portable security on a Roman frontier, but this cannot be established until a
corpus of the material is finalized.
130
James and Thorpe, 273.
131
J. W. Brailsford. Guide to the Antiquities of Roman Britain. (London: The Trustees of the
British Museum, 1966), 76.
132
McWhirr, 48.
85
The Padlock Form
Casing
Padlocks do not follow a standard form for their design and appear to be
constructed as bespoke objects. Though irregular padlocks do occur in anthropomorphic
and zoomorphic forms, the casing of the padlock is commonly cylindrical or square. (See
Fig. 24) The casing functions in much the same way as the installed lock form, housing
the functional components of the lock. Anthropomorphic and zoomorphic designs
certainly indicate a distinctive quality of their function along a social nature and are
padlocks which are likely meant to be used in public places. (See Figs. 25 and 26) Such
devices are bespoke objects, commissioned by those wealthy enough to afford a
decorative locking mechanism. Generic padlocks of a cylindrical or square nature serve a
purpose for everyday objects and the mundane, more utilitarian in their purpose.
The casing will have a lockplate, though it will typically not be the same form of
plate as those found on standard locks. The standard lockplate of padlocks will have a
square shaped keyway for the insertion of a key designed specifically for the function of
the barb. It is important to note that the compact structure of rotary locks allowed for
their function to be used in padlocks along with wards. A rotary key survives within the
body of a discarded padlock in the British Museum archives in Figure 27. The
rectangular or irregular keyway is an indicator of a rotary lock and survives on the
lockplates found on surviving padlocks recovered from Sardis.133 Such locks can be
133
Jane C. Waldbaum. Metalwork from Sardis: The Finds through 1974. (London and
Cambridge: Harvard University Press, 1983), 69-78.
86
distinguished as rotary padlocks, but only after proper analysis confirms them as such.
The general term “padlock” is sufficient until at such a time the locking technology is
known, as most padlocks use compression technology in the use of barbs within the
casing of the mechanism.
Retainer and Barb
The first padlocks utilized a unique design and function that corresponds with
modern versions. A retaining bar, or pin, would be inserted into the body of the padlock.
The retainer would be spliced at the end with a small flaps bent backwards towards the
body of the retainer. This created barbs which functioned as springs. A barb acts as the
component of the lock which actually secures the device. As the retainer is pushed into
the face of the lock body, the barbs compress down. The barbs spring open once they
pass the threshold of the lock body, preventing the retainer from being removed, save
with the use of a key. To better understand this function, Figure 28 shows a cut-away
view of a standard Roman padlock to reveal how such a design operates.
The retainer is used to secure the desired object, which may have often been
sections of chain. Padlocks are used in the Roman period in much the same way that they
are used today. Such devices likely secured chains, or perhaps a soldier’s footlocker, but
for the first time, security was portable and reusable wherever the owner desired. The
use of padlocks along with chains was common, as the iron padlock recovered from
Fishbourne palace was recovered in situ with a chain.134 This padlock can be granted a
definitive range of date for its utility, as Fishbourne palace was destroyed in a fire and the
87
pottery recovered from the excavation dates between 270-296 CE.135 Padlocks survive
not only at Fishbourne Palace, but also Lullingstone Roman villa.136 The padlock at
Lullingstone was an iron padlock, found along with other various iron objects, including
a chain.137
Though chain is a common medium for securing objects, the retainer could
theoretically be used on a variety of devices, such as jewelry boxes, chests, and anything
to which the retainer could fit. Even uncommon objects incorporate the technology into
devices. One example is of handcuffs recovered from Silchester which utilize the
padlock function in securing the device.138 As the shaped iron rings were closed around
the wrists of the restrained, the barbs at the end would insert into a padlock body.
Though the form is different, the action of the barbs is the same for other padlocks which
utilize such technology and shows the versatility that such locking mechanisms provided
in the Roman world. Shackles and manacles are likely used in varying degree in the
Roman period for the restraint and transport of slaves and criminals.
Compression Key Form
________________________
134
Barry Cunliffe. Fishbourne Roman Palace. (Stroud: Tempus Publishing, Ltd., 1999), 141.
135
Ibid., 142.
136
Bédoyère, 114.
137
G. W. Meates. Lullingstone Roman Villa: Ministry of Public Building and Works Guide
Book. (London: Her Majesty’s Stationery Office, 1967), 41.
138
Miranda Green. Aspects of Roman Life: Roman Technology and Crafts. (Essex: Longman
Group, Ltd., 1982), 42.
88
The keys utilized for the padlock do not often survive with the context of the
original padlock, save in Figure 27, where the key was discarded with the device. A
reconstruction of the key is possible if the retaining bar and barb form is known, as the
action of the key is simply an insertion of the key over the barbs to compress them and
remove the retainer. Such keys are unique from the tumbler and rotary forms though they
maintain the same three sections of anatomy. The key may have a handle to be carried on
a leather thong or metal ring, a shank that follows the length of the body for insertion into
the padlock, and a bit that matches the lockplate and barbs for the casing of the lock. In
much the same way with tumbler technology, the intuitive design of the padlock can
allow for a general reconstruction of the key. The length of the shank follows the length
of the casing of the padlock, as the key would allow the retainer to leave the threshold
through which it was inserted, and additionally, the structure and size of the bit can be
estimated if the barbs or the lockplate remain.
Rotary Key Form
Because rotary technology provides a compact structure for installed rotary locks,
the same lock technology and key form can be used on the padlock. The shape, length,
and use of perforations, grooves, and channels can occur with rotary keys for padlocks.
The distinguishing characteristic of a padlock which uses a rotary key will still be found
in the lockplate, as the rectangular or irregular keyway is used for the insertion of rotary
89
keys. Though rotary keys are utilized, preference is given to the compression form due to
its simplistic yet effective design.139
The components of a rotary lock would necessitate a variation in the internal
design of the padlock, as compression technology utilizing barbs and retainers is selfactuated. The security component is a part of the retainer itself, operated by a
compression key. A rotary padlock relies on an internal hasp or spring plate that would
secure itself onto the retainer when it is closed and locked, being removed by the action
of the key and therefore freeing the retainer. Though not visible in this instance, the
padlock in Figure 24 displays this concept in the retainer, as it does not have the
compression barbs, but rather a grooved recess.
The choice to use a compression padlock over a rotary padlock is unknown,
though it may be due to a security design flaw that all rotary locks share. Rotary locks
rely on the free action of the internal spring or deadbolt for the locking and unlocking of
any of the devices that follow this technology, but that also implies that any object can be
crafted as a lockpick to fit into the lock body and manipulate the deadbolt or hasp.
Compression technology utilizes multiple barbs, increasing the difficulty of manipulation
if all barbs need to be compressed at the same time to remove the retainer. Such devices
are effective in their operation and simplistic in their construction, which may explain the
preference for such padlocks in the surviving materials.
139
James and Thorpe, 472.
90
Fig. 24: Barrel-Shaped Padlock
Access: # X 490.475 Barrel Shaped Padlock
Photographer: Ursula Pariser
Courtesy of the Menil Collection, Houston, Texas
Fig. 25: Byzantine Padlock in the form of a Bull
Access: # X 490.697 Mid-Byzantine Padlock in the
form of a Bull
Photographer: Ursula Pariser
Courtesy of the Menil Collection, Houston, Texas
91
Fig. 26: Anthropomorphic Padlocks
Image AN921585001
Courtesy of the Trustees of the British Museum
Fig. 27: Rotary Key with Padlock
Image AN372210001
Courtesy of the Trustees of the British Museum
92
Fig. 28: Reconstructed Roman Padlock
Image AN192085001
Courtesy of the Trustees of the British Museum
93
CHAPTER 8
PHYSICAL CONTEXT
Locks and keys are used in a variety of physical contexts, each structured for the
object it secures. Chests and cabinets undoubtedly utilized smaller locks and keys, as the
containers are relatively smaller than doorways. But such devices can maintain physical
allure, as the materials and construction relay a social status in their physical context.
The presence of a lock relays a sense of hidden affluence, as the owner can display a
highly ornate lock as a means of allusion to the wealth preserved inside. This does not
have to apply strictly to physical locks. A stone chest from the British Museum even
retains the form of a tumbler lockplate chiseled upon the face, implying a standard of
social acknowledgment for such devices. (See Fig. 29)
Chests and Cabinets
Though he foregoes social implications, Guy de la Bedoyere observes the use of
ring keys for chests and cabinets.140 Ring keys are compact and clearly operate smaller
locks commonly found on small jewelry chests, as opposed to doors. Hundreds of
fittings for furniture and chests have been recovered from Sardis and their dates range
from Late Roman to Early Byzantine.141 The objects range from plain to decorative and
were secured in much the same way that doors and entryways are secured with only a few
140
Bédoyère, 124.
141
Waldbaum, 69-78.
94
exceptions. Chests and cabinets often utilized hinges and lids for opening and required
the use of a hasp to secure the lid to the lock body, as described in chapter 2. Two of the
lockplates recovered from the Tal-y-Llyn Roman hoard have the distinctive recess for the
hasp on the face of the lockplate.142
Doors and Gates
Locks could be affixed behind a door, or within it, much like our modern locking
devices. The hardware would therefore be concealed, not visible to the eye. Roman
entryways do not always have a single pane doorway mounted on the threshold, but two
or more, based on the surviving entryways in public and private space at Pompeii.143 The
deadbolt for such cases would likely secure the two doors together. In the instance of a
single door, the deadbolt would be thrown most commonly into the jamb of the door
frame for rotary and tumbler keys, though the lift key form would pull a deadbolt out of
the threshold or lentil of the door.
Since gates provide the first and immediate means of entry into a city or palace,
they are the most vulnerable to assault and siege. Gates would be barred with large
timbers in the event of an attack by a foreign force. But gates could also utilize locks for
additional security, as is evidenced in classical literature. In confronting the suitors of his
household, Odysseus orders Philoetius to “shoot the bolt of the courtyard’s outer gate,
lock it, and lash it fast.144” Gates could certainly be secured by lock and key, as is
evident in the contingencies formed for the purpose of dismantling a gate during siege in
142
Spratling, 229.
143
Roberto Cassanelli, et al. Houses and Monuments of Pompeii: The Works of Fausto and
Felice Niccolini. (Los Angeles: The J. Paul Getty Museum, 2002), 84, 112.
95
which the hinges and bars are sawn through to bring the gate down.145
Keys used to operate doors and gates vary in a stylistic manner within the Levant.
The handle of the key is replaced by a rounded knob, which may have allowed the key to
be worn at the waist with the use of a metal clip or hoop attached to the belt.146 Though
the style is different from the Roman form commonly seen, the function of the keys and
the technology for which they are used is introduced with the occupation of the Levant
under Roman rule. The technology of the rotary lock and perhaps some aspect of the
tumbler lock may be borrowed from Roman influence.147
Keys to the City
City gates, being locked by larger mechanisms similar to those used on the
domicile, utilized a key that would have been held by the ruler or even public officials.
One of the distinguishing characteristics of a city’s surrender and subjugation is the
relinquishment of the keys to the city upon the enemy.148 Granting the keys to the enemy
could be done as a symbol of entrust or surrender, but was also utilized to control the
people within. In The History of Rome, Livy makes mention of an attempt to quell a
revolution by removing the keys from the magistrates, then locking and guarding the
gates to the city.149 The keys themselves may have not been duplicates, but rather
________________________
144
Homer, 21.267-269.
145
Polybius Histories. 7.16.5-6.
146
Yadin, 94-100.
147
Marti Lu Allen. “’The Keys of the Kingdom’: Keys from Masada,” in Masada and the
World of the New Testament, ed. John Franklin Hall (Utah: Brigham Young University, 1996), 166.
148
Herodotus The Histories. 3.155.6.
149
Titus Livius The History of Rome. 27.24.8.
96
singular keys that operated a single gate. A collection of city gate keys from 17th century
Milan display qualities showing they are for different gates.150 Key bearers throughout
the centuries likely had power over one gate, which also ensures that no single person has
power over all the gates within the city.
Guilds
Artisans maintaining a common craft are known to congregate together under
representative guilds through various periods. Epigraphic evidence survives in the
Roman Period listing such guilds as collegia, and was formed for religious purposes
along with the social establishment of craftsmen.151 The primary concern of the collegia
was the stability of the community and the maintaining of social ordinances and rites
established by the group. No direct evidence exists of a collegium for locksmiths, but the
concept was not uncommon for crafts and such a guild likely existed in different urban
contexts where a variety of metalworkers were needed to meet the needs of the people
and the Empire.
At minimum there exists evidence for the presence of metalworkers who
specialized as locksmiths. The materials recovered from Knossos bear resemblance in
their construction to one another in the lock parts preserved in the cave burials.152 The
similarities in the components suggest that the same individual constructed each of the
locks, but specialized in securing small chests and furniture.153 In addition, hundreds of
150
Lise, 50-1.
151
John Bodel. Epigraphic Evidence: Ancient History from Inscriptions. (London and New
York: Routledge, 2001), 116.
152
Wardle, 477.
153
Ibid.
97
locking components have been recovered from ancient Sardis, dating from the Late
Roman to the Early Byzantine. The recovered objects are comprised of padlocks,
lockplates, and keys, as well as hinges and furniture fittings common for chests.154 A
locksmith would be endemic in highly populated areas where security is necessitated.
The construction and repair of security devices, as well as replacing broken keys would
need to be addressed regularly.
One rotary key bears uncanny resemblance to the rotary key ring from Figure 23
of the Menil Collection.155 At first glance, the key appears as a duplicate, but possesses
an extra incision in addition to a different placement for the channel. The provenance is
not listed for the key, but is undoubtedly made from the same casting as the Menil key
with alterations made to the bit for a different lock, hinting at the possibility of a common
craftsman.
Anthropomorphism and Zoomorphism
Artifacts constructed with zoomorphic and anthropomorphic design are not
uncommon in the Roman period, nor are they uncommon amongst locks and keys.
Roman constructions of chests rely consistently on motifs and scenes of mythological
significance for aesthetic appeal. A variety of examples can be found in Aladar
Radnoti’s publication, Möbel und Kästchenbeschläge, Schlösser und Schlüssel. Scenes
from the Greco-Roman mythos and various gods and goddesses and their feats adorn a
variety of lockplates and the chest decorations surrounding the lockplates.156 But it was
154
Waldbaum, 69-79.
155
Eras, 34.
156
Aladar Radnoti. “Möbel und Kästchenbeschläge, Schlösser und Schlüssel,” in Geschichte
98
far more common amongst the Romans to construct locks and keys in a zoomorphic or
anthropomorphic fashion.
Zoomorphic Keys and Locks
In the anatomy of the key, the handle is often replicated into the body or shape of
an animal. Such items are generally made from copper-alloy, as such metal can be cast
from a mold. Such handles could then be fit to a shank and bit of a different or same
metal. (See Fig. 30) A variety of zoomorphic handles can be found in the Metropolitan
Museum of Art archives which display the range of animal types commonly cast for
keys. A ram and a horse head key handle can be found (See Figs. 31 and 32), but the
bounding lion appears to be the common form. (See Figs. 33, 34, and 35) The use of
zoomorphic designs as lock forms as opposed to motifs may begin to occur in the
Byzantine period, as latches and lockplates take the form of lions at Byzantine
Pergamon.157
Anthropomorphic Keys and Locks
Human attributes for the handle of the key was not uncommon. Several Roman
key handles can be found in the Metropolitan Museum of Art displaying
anthropomorphic qualities. The common form appears to be a single arm clutching an
object. (See Fig. 36) The human head also occurs, though one particular handle has the
entire human form on display, perhaps in the form of an orator making a defense. (See
Figs. 37 and 38) Such objects may carry a social weight that the zoomorphic figures do
________________________
der Stadt in der Römerzeit eds. Maria Alföldi and Laszlo Barkoczi (Budapest: Akad miai Kiad , 1957),
241-363.
157
Machteld J. Mellink. “Archaeology in Anatolia,” American Journal of Archaeology 93
(1989): 125, 128.
99
not display and may be an indication of a commissioned object, perhaps for an office or
individual of prestige.
Anthropomorphism amongst the components of the lock may not have been
common, except in the case of padlocks. Late Roman padlocks appear to have
incorporated human faces as a portion of padlocks body and interface. (See Fig. 26) The
use of anthropomorphism for locking components seems rare and tends to be found more
often on keys. Constructions of locking components focus more on motifs and display,
such as the Roman lockplates found in Radnoti’s publication and the Byzantine padlock
from the Menil Collection Archives. (See Fig. 39)
Deity
The display of a deity is irregular in keys. Though we know of keys’ association
with the divine, as in the case of the Homeric key belonging to the temple of Artemis at
Lusoi, anthropomorphic design which reveals a deity appears on only one known
occasion. Heinrich Schliemann’s excavation of Troy yielded a single key in particular
that may have been purely decorative, but is nonetheless unique. An anthropomorphic
figure adorns a bronze key of seeming rotary form.158 The head appears at the end of the
key, but the shaft of the key has stumps on the sides for arms, as well as genitalia on the
inverse of the shaft. The presence of a phallus on the body of the key is characteristic of
Hermes and the Roman counterpart, Mercury.159
158
Schliemann. Ilios. 620.
159
Ibid., 621.
100
Fig. 29: Stone Chest Inscribed with Tumbler Lockplate Form
Image AN975174001
Courtesy of the Trustees of the British Museum
Fig. 30: Ram Head Key Handle
Access. # 06.176.38
Copper-Alloy Roman Key Handle, Iron Shank
1st- 7th Century CE
Courtesy of the Metropolitan Museum of Art
101
Fig. 31: Horse Head Key Handle
Access. # 06.176.24
Copper-Alloy 2nd-3rd Century CE
Courtesy of the Metropolitan Museum of Art
Fig. 32: Ram Head Key Handle
Access. 06.176.41
Copper-Alloy
1st- 7th Century CE
Courtesy of the Metropolitan Museum of Art
102
Fig. 33: Key Handle in the form of a Lion
Access. 06.176.48
Copper-Alloy Roman Key Handle
1st- 7th Century CE
Courtesy of the Metropolitan Museum of Art
Fig. 34: Bounding Lioness Key Handle
Access. # 06.176.34
Copper-Alloy
1st- 7th Century CE
Courtesy of the Metropolitan Museum of Art
103
Fig. 35: Bounding Lion Key Handle
Access. # 06.176.35
Copper-Alloy, Iron Shank
1st- 7th Century CE
Courtesy of the Metropolitan Museum of Art
Fig. 36: Key Handle in the Shape of an Arm, Clutching an Object
Access. # 06.176.21
Copper-Alloy
1st- 7th Century
Courtesy of the Metropolitan Museum of Art
104
Fig. 37: Key Handle in the Shape of a Human Head
Access. # 06.176.18
Copper-Alloy
1st- 7th Century
Courtesy of the Metropolitan Museum of Art
Fig. 38: Key Handle of an Orator
Access. # 06.176.74
Copper-Alloy
1st- 2nd Century
Courtesy of the Metropolitan Museum of Art
105
Fig. 39: Middle Byzantine Spring Padlock
Acces. # X 490.474
Photographer: Ursula Pariser
Courtesy of the Menil Collection, Houston, Texas
106
CHAPTER 9
SUMMARY AND CONCLUSION
Summary
After examining materials available from a variety of major collections as well as
examples found within scholarly publication, it can be concluded that definitive lock and
key forms exist in the ancient world. Evaluation of these materials has shown that a clear
evolution of form has been used in the Roman period, some of which has influenced the
technology of subsequent periods. The Metropolitan Museum of Art, The Menil
Collection, The Deutsches Museum, and The British Museum each contain variations of
these lock or key forms, in addition to their aesthetic attributes.
To summarize, locks can be broken down into three different categories of
technology. The Tumbler Form, Rotary Form, and Padlock Form each comprise a clear
demarcation of technological change indicated by physical diagnostics. Due to the
precise nature of the technology, the keys created for their use display form variation.
The forms associated with the tumbler form are Laconian, Slide-Key, and Lift Form. Of
these key types, the laconian is likely the rarest, being the remnant of a key form
predating the Roman period. The slide-key and lift form took the place of the laconian
key, adding complexity and difficulty to the tumbler lock to prevent unwanted
infiltration.
The Rotary Form of lock technology stands out as a transitional phase for Roman
107
lock technology. Being incorporated into the repertoire of lock use some time just before
or during the Roman period, it does not become the popular form until the Byzantine
period. Tumbler technology is phased out at this time, and rotary forms are then utilized
from the Byzantine period until the Industrial Revolution. The key forms utilized follow
one of two types. The Grooved Rotary Key and the Perforated Rotary Key each show
distinct diagnostics in their structure and creation which denote the form of the lock in
which they are used. In use, the grooved rotary key appears in a significantly lower
frequency to the perforated rotary key in the comparative data currently available, but
assertions regarding the proclivity of their use cannot be made until a corpus of the
material is established.
The Padlock abides in the record as one of the more significant contributions to
lock technology which may have been endemic to Roman technological advancement. It
is difficult to establish precedence for padlocks as a Roman invention, as padlocks come
into use in East Asia within a century of the first Roman padlocks. The form is different
though the technology follows the same pattern of use. If more information becomes
available, or if Roman padlocks continue to show an establish frame of use before the
earliest Asiatic padlock forms, then the invention may be attributed to the Romans. Lock
forms for padlocks can hold rotary technology, but also show a dispensation towards
compression technology. Keys for padlocks will be in either the Compression Key form,
though the level to which rotary keys have been used on such locks is unknown.
Keys appear as rings to be worn, holding a functional use that is both aesthetically
and socially purposeful. The social implications of anthropomorphic and zoomorphic
design add a layer of anthropological evaluation to bespoke objects recovered in
108
archaeological excavation. These designs, coupled with their status as seen in Ancient
texts, speaks of a cultural significance locks and keys held as symbols of power and
authority.
Conclusion
The social background of locks and keys has revealed significant information for
providing a typology of the materials, but it has also revealed much of the
anthropological background of the culture that developed them. The role that locks and
keys play in everyday life cannot be overlooked. Such materials are more than just a
physical barrier in the ancient world, but also serve as a socio-cultural identifier and a
means of conveying personal wealth and affluence. The role of such materials was
significant within a religious and social context and warrants further examination of the
materials in the chora of personal and public use for an evaluation of religious practice
and criminology.
The typology created can at last place locks and keys within collections into an
established taxonomy. In the case of keys on display, they will now illuminate the lock
form for which they were used, allowing collections and museums further
acknowledgment to the object’s technological origin and use. But such a typology will
not only benefit public collections, but also field methodology in archaeology.
Access Analysis allows for an evaluation of public and private space based on the
surviving architecture, but evidence of security within the architecture or in the body of
recovered materials will contribute additional information to an evaluation of space. The
keys recovered in the field will allude to the typology of the lock, and therefore both the
complexity and utility of the space it guarded. A complex key with multiple tines may
109
not necessarily mean that what it secured was culturally or economically superior to
another chora, but the presence of decoration, anthropomorphism, zoomorphism, or even
crude simplicity may belie social implications for such an analysis. Applying the
typological format provided to distinguish an object’s form will benefit the known
security of a passageway. Points of access which possess known security will provide an
addition to anthropological analysis, adding the questions of “Who had access?” and
“What was being secured?” to the litany of evaluation. Archaeology is a growing field,
constantly in flux as new information surfaces, and in the case of this typology, a corpus
of the material will provide a new avenue for anthropology and archaeology. The next
phase of research following such a typology should be the development of that corpus, to
provide a complete listing of the known artifacts according to the typology.
110
BIBLIOGRAPHY
Aeschylus. Euminides.
Allen, Marti Lu. “’The Keys of the Kingdom’: Keys from Masada.” In Masada and the
World of the New Testament, ed. John Franklin Hall. Vol 36, Number 3. Utah:
Brigham Young University, 1996-1997.
Apuleius. Metamorphoses.
Aristophanes. Thesmophoriazusae.
Augustine of Hippo. De Doctrina Christiana. In Corpus Christianorum: Series Latina.
Turnholti: Typographi Brepols Editores Pontificii, 1962.
de la Bédoyère, Guy. The Finds of Roman Britain. London: B.T. Batsford Ltd., 1989.
Bodel, John. Epigraphic Evidence: Ancient History from Inscriptions. London and New
York: Routledge, 2001.
Brailsford, J.W. Guide to the Antiquities of Roman Britain. London: The Trustees of the
British Museum, 1966.
Brooke, W.P. ed. The Jugurtha of Sallust. London: Rivingtons, 1885.
Buchhoelz, Hans Guenter. Ugarit, Zypern und Aegaeis: Kulturbeziehungen in Zweiten
Jahrtausend v. Chr. Band 261. Muenster: Ugarit-Verlag, 1991.
Caple, Chris. Objects: Reluctant Witnesses to the Past. London and New York:
Routledge, 2008.
Cassanelli, Roberto, Pier Luigi Ciapperelli, Enrico Colle, and David Massimiliano.
Houses and Monuments of Pompeii: The Works of Fausto and Felice Niccolini. Los
Angeles: The J. Paul Getty Museum, 2002.
Chrysostom, John. On Virginity.
Cicero, M. Tullius. De Natura Deorum.
______________. Pro Marcus Tullius.
______________. The Fourteen Orations against Marcus Antonius.
111
Clark, Gillian. Women in Late Antiquity: Pagan and Christian Lifestyles. Oxford:
Oxford University Press, 1994.
Clement of Alexandria. Paedogogus.
Cunliffe, Barry. Fishbourne Roman Palace. Stroud, Gloucestershire: Tempus Publishing,
Ltd., 1999.
Curtis, John and Matthew Ponting. An Examination of Late Assyrian Metalwork: With
Special Reference to Nimrud. Oxford: Oxbow Books, 2013.
Darvill, Timothy. “Latchlifter” In, The Concise Oxford Dictionary of Archaeology. 2nd
Edition. Edited by Timothy Darvill. Oxford: Oxford University Press, 2008.
DeMarrais, Elizabeth, Chris Gosden and Colin Renfrew, Eds. Rethinking Materiality: the
Engagement of the Mind with the Material World. Cambridge: McDonald Institute
for Archaeological Research, 2004.
Eras, Vincent J.M. Locks and Keys throughout the Ages. New York: Lip’s Safe and Lock
Manufacturing Company, 1957.
Euripides. Medea.
Eustathius of Thessalonica. Commentary on the Odyssey.
Finkelstein, Louis, Saul Lieberman and Yigael Yadin. Masada and the Finds from the
Bar-Kokhba Caves. New York: The Jewish Museum, October 1967.
Gardner, Jane F. Women in Roman Law and Society. Bloomington and Indianapolis:
Indiana University Press. 1986.
Gellius, Aeulus. Noctes Atticae.
Green, Miranda. Aspects of Roman Life: Roman Technology and Crafts. Essex:
Longman Group Ltd., 1982.
Haensch, Rudolf. “Inscriptions as Sources of Knowledge for Religions and Cults in the
Roman World of Imperial Times.” In, A Companion to Roman Religion. Jörg
Rüpke, Ed. Malden, MA: Blackwell Publishing, 2009.
Haimovich-Carmin, Noga, Edward Levin, Carl Ebert, et al. “Churches and Monasteries
in Samaria and Northern Judea.” In Christians and Christianity, Vol. III. JSP. Israel
Antiquities Authority, 2012.
Healy, J. F. Mining and Metallurgy in the Greek and Roman World. London: Thames
and Hudson, 1978.
Herodotus. The Histories.
112
Homer. The Odyssey.
Hurcombe, L. M. Archaeological Artifacts as Material Culture. New York: Routledge,
2007.
Husselman, Elinor M. Karanis Excavations of the University of Michigan in Egypt 19281935: Topography and Architecture. Ann Arbor: The University of Michigan
Press, 1979.
James, Peter and Nick Thorpe. “Locks and Keys”. In Ancient Inventions. New York:
Ballantine Books, 1994.
King, Philip J. and Lawrence Stager. Life in Biblical Israel. London: Westminster John
Knox Press, 2001.
Kirwan, Lawrence, P. The Oxford University Excavations at Firka. Oxford and London:
Oxford University Press, 1939.
Kuefler, Mathew. “The Marriage Revolution in Late Antiquity: The Theodosian Code
and Later Roman Marriage Law.” Journal of Family History 32 (2007): 343-55.
Lise, Giorgio. Chiavi e Serrature. Milan: BA-MA Editrice, 1987.
Livius, Titus. The History of Rome.
Made of Iron. September-December 1966. Houston, Texas: University of St. Thomas Art
Department and The Menil Collection, 1966.
Martial. Epigrammata.
Marquardt, Joachim. Römische Staatsverwaltung. Vol III. New York: Arno Press, 1975.
Marquardt, Joachim and Theodor Mommsen. Handbuch der Römischen Alterthümer.
Sechster Band. Leipzig: Verlag von S. Hirzel, 1885.
McWhirr, Alan. Roman Crafts and Industries. Aylesbury: Shire Publications, Ltd.,
1982.
Meates, G.W. Lullingstone Roman Villa: Ministry of Public Building and Works Guide
Book. London: Her Majesty’s Stationery Office, 1967.
Mellink, Machteld J. “Archaeology in Anatolia.” American Journal of Archaeology. 93:
1 (Jan. 1989): 105-33.
Müller, Allan Clarence. “The Archeology of the Dorians.” Ph.D. diss., University of
California, March 1970.
Neuburger, Albert. Die Technik des Altertums. New York: Barnes and Noble, 1930.
113
Ovid. Ars Amatoria.
____. Fasti.
Papaconstantinou, Demetra. Deconstructing Context: a Critical Approach to
Archaeological Practice. Oxford: Oxbow Books, 2006.
P., F.N. “Keys and Locks”. M Cary, et al. eds. The Oxford Classical Dictionary. Oxford:
Clarendon Press, 1949.
Paton, W.R. Trans. Polybius: the Histories. Volume III. London: William Heinemann
Ltd, 1972.
Philip, Graham. Tell El-Daba XV: Metalworking and Metalworking Evidence of the Late
Middle Kingdom and the Second Intermediate Period. Vienna: Verlag der
Oesterreichischen Akademie der Wissenschaften, 2006.
Plautus, Titus Maccius. Mostellaria.
Pliny the Elder. Naturalis Historia.
Plutarch. Lives.
Polybius. Histories.
Radnoti, Aladar. “Möbel und Kästchenbeschläge, Schlösser und Schlüssel.” Maria
Alföldi, Laszlo Barkoczi, et al., eds. In, Geschichte der Stadt in der Römerzeit.
Budapest: Akad miai Kiad Archaeologica Hungaria, 1957.
Read, Dwight W. Artifact Classification: A Conceptual and Methodological Approach.
Walnut Creek: Left Coast Press, 2007.
Robinson, O.F. The Criminal Law of Ancient Rome. Baltimore: The Johns Hopkins
University Press, 1995.
Roche, Paul. Trans. Three Plays of Euripides: Alcestis, Medea, The Bacchae. New York
and London: W.W. Norton Company, 1974.
Rouse, Irving. “The Classification of Artifacts in Archaeology,” American Antiquity 25:3
(Jan. 1960): 313-23.
Sallust. Bellum Jugurthinum.
Salonen, Armas. Die Tueren des Alten Mesopotamien: Eine Lexikalische und
Kulturgeschichtliche Untersuchung. Helsinki: Suomalaisen Kirjallisuuden
Kirijapaino Oy, 1961.
Schliemann, Heinrich. Ilios: City and Country of the Trojans. New York: Harper and
Brothers, Franklin Square, 1881.
114
_______________. Troy and Its Remains: A Narrative of Researches and Discoveries
Made on the Site of Ilium and in the Trojan Plain. London: John Murray,
Albemarle Street, 1875.
Sellers, O.R. and D.C. Baramki. “A Roman-Byzantine Burial Cave in Northern
Palestine.” ASOR 15-16 (1953): 7-55.
Seneca. De Ira.
Singer, Charles, E.J. Holmyard, A.R. Hall, and Trevor I Williams, eds. A History of
Technology. Vol. II. Oxford: Clarendon Press, 1954.
Smith, Sir William. A Dictionary of Greek and Roman Antiquities. Vol. 1. London: John
Murray, 1901.
Spratling, Mansel, G. “The Date of the Tal-y-Llyn Hoard.” Antiquity 45: 159 (Sept-Oct.
1966): 228-9.
The Mishnah. Danby, Herbert., trans. London: Oxford University Press, 1933.
The Tosefta: Second Divison, Moed. Jacob Neusner, trans. New York: Ktav Publishing
House Inc., 1981.
Titus Maccius Plautus. Mostellaria.
Varro. De Lingua Latina.
Vikan, Gary and John Nesbitt. Security in Byzantium: Locking, Sealing, Weighing.
Washington D.C: Dumbarton Oaks, 1980.
Vikan, Gary. “Unpublished Chapter on Keys and Locks from a Book on the Byzantine
Small Finds in the Menil Foundation Collection.” Houston: Menil Collection
Archives, 1994.
Virgil. The Aeneid.
Waldbaum, Jane C. Metalwork from Sardis: The Finds through 1974. London and
Cambridge: Harvard University Press, 1983.
Wardle, Diana and K.A. Wardle. “Glimpses of Private Life: Roman Rock Cut Tombs of
the First and Second Centuries AD at Knossos.” In, Knossos, Palace, City, State.
Gerald Cadogan, Eleni Hatzaki, and Aonis Vasilakis, eds. London: British School
at Athens Studies 12, 2004.
Wilkinson, J. Gardner. The Ancient Egyptians: Their Life and Customs. Volumes I-II.
London: John Murray Publishing. 1994.
Yadin, Yigael. The Finds from the Bar-Kokhba Period in the Cave of Letters. Jerusalem:
Israel Exploration Society, 1963.