Computerassisted Query Formulation Alvin Cheung

Armando Solarlezama download

https://ebookbell.com/product/computerassisted-query-formulation-
alvin-cheung-armando-solarlezama-57526902

Explore and download more ebooks at ebookbell.com

Here are some recommended products that we believe you will be
interested in. You can click the link to download.

Computerassisted Language Learning Concepts Methodologies Tools And

Applications 4 Volume Set Management Association

https://ebookbell.com/product/computerassisted-language-learning-
concepts-methodologies-tools-and-applications-4-volume-set-management-
association-47605128

Computer Assisted Research On The Bible In The 21st Century Luis Vegas
Montaner

https://ebookbell.com/product/computer-assisted-research-on-the-bible-
in-the-21st-century-luis-vegas-montaner-49450366

Computer Assisted Research On The Bible In The 21st Century Luis Vegas
Montaner Guadalupe Seijas De Los Roszarzosa Javier Del Barco Editors

https://ebookbell.com/product/computer-assisted-research-on-the-bible-
in-the-21st-century-luis-vegas-montaner-guadalupe-seijas-de-los-
roszarzosa-javier-del-barco-editors-50343254

Computer Assisted Music And Dramatics Possibilities And Challenges

Ambuja Salgaonkar

https://ebookbell.com/product/computer-assisted-music-and-dramatics-
possibilities-and-challenges-ambuja-salgaonkar-50591942
Computerassisted Language Learning Learners Teachers And Tools 1st
Edition Jeongbae Son

https://ebookbell.com/product/computerassisted-language-learning-
learners-teachers-and-tools-1st-edition-jeongbae-son-51307738

Computerassisted Literary Translation Andrew Rothwell Andy Way

https://ebookbell.com/product/computerassisted-literary-translation-
andrew-rothwell-andy-way-56821570

Computer Assisted Optimization Of Cardiac Resynchronization Therapy

Raz Miri

https://ebookbell.com/product/computer-assisted-optimization-of-
cardiac-resynchronization-therapy-raz-miri-4099286

Computer Assisted Exercises And Training A Reference Guide Erdal

Ayirci

https://ebookbell.com/product/computer-assisted-exercises-and-
training-a-reference-guide-erdal-ayirci-4302292

Computer Assisted Assessment Research Into Eassessment International

Conference Caa 2014 Zeist The Netherlands June 30 July 1 2014
Proceedings 1st Edition Marco Kalz

https://ebookbell.com/product/computer-assisted-assessment-research-
into-eassessment-international-conference-caa-2014-zeist-the-
netherlands-june-30-july-1-2014-proceedings-1st-edition-marco-
kalz-4697446
Foundations and Trends® in Programming Languages
Vol. 3, No. 1 (2016) 1–94
© 2016 A. Cheung and A. Solar-Lezama
DOI: 10.1561/2500000018

Computer-Assisted Query Formulation

Alvin Cheung Armando Solar-Lezama

University of Washington MIT CSAIL
[email protected] [email protected]
 Contents

1 Introduction 2

2 Query Processing 5
2.1 Relational DBMS and Query Languages . . . . . . . . . . 5
2.2 DBMS as a library . . . . . . . . . . . . . . . . . . . . . . 6
2.3 The ORM approach . . . . . . . . . . . . . . . . . . . . . 7
2.4 Query Execution . . . . . . . . . . . . . . . . . . . . . . . 8

3 Program Synthesis 13
3.1 The Problem . . . . . . . . . . . . . . . . . . . . . . . . . 13
3.2 Deductive Synthesis . . . . . . . . . . . . . . . . . . . . . 14
3.3 Inductive Synthesis . . . . . . . . . . . . . . . . . . . . . 15

4 Using Verified Lifting to Rewrite Code into SQL 19

4.1 Interacting with the DBMS . . . . . . . . . . . . . . . . . 20
4.2 QBS Overview . . . . . . . . . . . . . . . . . . . . . . . . 22
4.3 Theory of Finite Ordered Relations . . . . . . . . . . . . . 29
4.4 Synthesis of Invariants and Postconditions . . . . . . . . . 39
4.5 Formal Validation and Source Transformation . . . . . . . 44
4.6 Preprocessing of Input Programs . . . . . . . . . . . . . . 46
4.7 Experiments . . . . . . . . . . . . . . . . . . . . . . . . . 49
4.8 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . 59

ii
 iii

5 Assisting Users Specify Database Queries 61

5.1 Intended Users . . . . . . . . . . . . . . . . . . . . . . . . 61
5.2 Usage Model . . . . . . . . . . . . . . . . . . . . . . . . . 64
5.3 Search Algorithms . . . . . . . . . . . . . . . . . . . . . . 68
5.4 Query Refinement . . . . . . . . . . . . . . . . . . . . . . 79

6 Conclusion and Future Work 83

6.1 Beyond Input-Output Examples . . . . . . . . . . . . . . . 83
6.2 Extending System Capabilities . . . . . . . . . . . . . . . 84
6.3 Refinement Techniques . . . . . . . . . . . . . . . . . . . 84
6.4 Combining Different Inference Algorithms . . . . . . . . . 85

References 87
Abstract

Database management systems (DBMS) typically provide an appli-

cation programming interface for users to issue queries using query
languages such as SQL. Many such languages were originally designed
for business data processing applications. While these applications are
still relevant, two other classes of applications have become important
users of data management systems: (a) web applications that issue
queries programmatically to the DBMS, and (b) data analytics involv-
ing complex queries that allow data scientists to better understand their
datasets. Unfortunately, existing query languages provided by database
management systems are often far from ideal for these application do-
mains.
In this tutorial, we describe a set of technologies that assist users in
specifying database queries for different application domains. The goal
of such systems is to bridge the gap between current query interfaces
provided by database management systems and the needs of different
usage scenarios that are not well served by existing query languages.
We discuss the different interaction modes that such systems provide
and the algorithms used to infer user queries. In particular, we focus
on a new class of systems built using program synthesis techniques,
and furthermore discuss opportunities in combining synthesis and other
methods used in prior systems to infer user queries.

A. Cheung and A. Solar-Lezama. Computer-Assisted Query Formulation.

Foundations and Trends® in Programming Languages, vol. 3, no. 1, pp. 1–94, 2016.
DOI: 10.1561/2500000018.
 1
Introduction

From financial transactions to online shopping, we interact with

database management systems (DBMSs) on a daily basis. Since the
initial development of relational database systems, various query lan-
guages such as SQL have been developed for users to interact with the
DBMS. Many of these languages proved very effective for what was
originally their primary application: business data processing (e.g., gen-
erating transaction reports at a financial institution). Unfortunately,
many important applications of DBMSs that have emerged in recent
decades have proven to be a less than ideal fit for the interaction models
supported by traditional DBMSs.
One particularly important extension to the business data process-
ing application space corresponds to applications with complex business
logic, such as social network websites, online shopping applications,
etc. Unfortunately, traditional query interfaces often make develop-
ing such applications difficult. First, the general-purpose languages in
which these applications are usually written (e.g., Java or Python) are
quite different from the query languages supported by the DBMS. This
forces developers to learn a new language—and often a new program-
ming paradigm altogether. For example, an application developer who

2
 3

is used to thinking about computation over objects stored in the pro-

gram heap will need to recast her computation in terms of structured
relations stored on disks when interacting with a DBMS. Moreover, in
addition to being concerned about efficient memory layout for retriev-
ing in-memory objects, she will also need to understand the costs associ-
ated with bringing objects into memory from the disk. This “impedance
mismatch” [Copeland and Maier, 1984] problem has plagued applica-
tion developers for decades. Today, this mismatch is often addressed by
application frameworks known as Object Relational Mapping (ORM)
Frameworks that eliminate the need to think in terms of two distinct
programming models. Unfortunately, the use of ORMs often imposes
significant performance costs [Subramanian].
There are many reasons for the performance cost of ORMs, but
one that is especially significant is that they encourage a programming
style where computation that could have been implemented with a sin-
gle query and a single round trip to the database is instead implemented
with several simpler queries connected together with imperative code
that manipulates their results. This is problematic because in addi-
tion to increasing the number of round trips and the amount of data
that needs to be transferred between the application and the DBMS,
doing so also increases the cost of the computation, since the DBMS
is in much better position to optimize queries compared to a general-
purpose code compiler trying to optimize a block of imperative code
that happens to implement a relational operation.
As an example, while a relational join between relations R and S
can be implemented using a nested loop, with each loop processing
tuples from the two respective relations fetched from the DBMS, it is
much more efficient to implement the join as a single SQL query, as
the DBMS can choose the best way to implement the join during query
optimization.
In this tutorial we focus on a new approach based on verified lift-
ing [Cheung et al., 2015] to reduce the performance cost of these ap-
plication frameworks, allowing programmers to enjoy the benefits of
the reduced impedance mismatch. The first step in this technique is to
identify places in the application code where the programmer is using
4 Introduction

imperative code to implement functionality that could be implemented

as part of a query. The second and most important step is to use pro-
gram synthesis technology to derive a query that is provably equivalent
to the imperative code. Once that is done, the third step involves gen-
erating a new version of the code that uses the query in place of the
original code.
The technology behind this work was originally published ear-
lier [Cheung et al., 2013]. In this paper, we expand on the content
of that original paper in order to make the technology more accessi-
ble to researchers without a strong background in program synthesis
or verification, as well as to researchers who may not be as familiar
with database concepts. In Section 2, we provide a quick primer on
query execution and query processing, focusing on key concepts that
will help the reader understand the reasons for the performance prob-
lems introduced by ORMs. Section 3 provides a comprehensive primer
on program synthesis technology, focusing in particular on the tech-
niques that are leveraged by QBS, and putting them in context of
other synthesis technologies. Section 4 describes the details of the QBS
approach, and finally Section 5 describes the state of the art in terms of
applications of synthesis to interact with DBMS systems and promising
directions for future work.
 2
Query Processing

This section provides a high-level introduction to the different ways

applications can interact with the DBMS. The section also provides
some background on how the DBMS processes queries; this background
will help the reader understand why issuing queries in different ways
can have significant performance impacts for an application.

2.1 Relational DBMS and Query Languages

Since the development of relational database systems in the 1970s,

SQL (Structure Query Language) has become the most popular query
language for interacting with DBMS. SQL is based on the relational
model, which models data as relation instances. A relation instance
is similar to a spreadsheet table with rows and columns, except that
columns are well-typed. Each column is a named and typed field, and
the set of fields for each relation is known as the schema of that relation.
An instance of a relation is a set of records (also called tuples), where all
records share the same schema. SQL is an implementation of relational
algebra [Codd, 1970, Date, 2000], except that it models relations as bags
(i.e., multisets) rather than sets, as in the original relational algebra

5
6 Query Processing

formulation. The language includes a subset of relational operators such

as projections, selections, joins, and aggregations, but not others such
as transitive closure. In addition to these operators, inLogical Plan ge
practice most DBMSs allow users to execute arbitrary code by defining
user defined functions (UDFs). To use UDFs, developers first implement
the UDF using a domain-specific language such as PL/SQL (which is an
implementation of the SQL/PSM standard [International Organization
for Standardization, 2011]) and compile it using a custom compiler
provided by the DBMS. The compiled binary is then linked to the
DBMS kernel and the function is then available to the query executor,
to be described in Section 2.4.

2.2 DBMS as a library

Since the initial development, relational DBMSs have been designed to

be stand-alone systems rather than application libraries to be linked
with the application during compilation. Early DBMS implementations
did not support complex applications and only provided a command-
line interface for end-users to interact with the system by typing queries
on the console, with the DBMS returning results and displaying them
to the user on the screen. As business data processing applications
became popular, DBMS implementations started to provide language
level abstractions (such as JDBC [JDBC 4.2 Expert Group, 2014] and
ODBC [International Organization for Standardization, 2008]) for ap-
plications to interact with DBMSs programmatically by issuing SQL
queries. Such abstractions are often implemented as connector libraries
provided by the DBMS, and are linked by developers to their applica-
tion binaries at compile time. These libraries allow application devel-
opers to embed query statements within their application source code
as if they were using the command-line interface, and are furthermore
completely separated from the DBMS implementation itself in order
to support functionality such as issuing queries remotely via network.
In these situations, the embedded query statements are sent to the
connector libraries as the application executes, which in turn are for-
warded to the DBMS for execution. The results are then sent back to
2.3. The ORM approach 7

the libraries, and the libraries would return them to the application
after serializing the results into data structures such as lists or arrays
of primitive types.
On the one hand, connector libraries greatly ease DBMS develop-
ment as they cleanly abstract away the application; the DBMS can pro-
cess queries as if they were issued by end-users through the command-
line interface. Unfortunately, this comes at a cost for the application
developer. First, as applications are usually written using a general-
purpose language, developers need to learn a new language in order to
express their persistent data needs. Worse yet, embedding query state-
ments as raw strings in the application makes debugging difficult; the
raw strings are not parsed or type-checked by the application compiler,
so errors in the queries only become apparent at execution time. Not
only that, embedding raw query strings in application code is often the
source of various security vulnerabilities such as SQL injection.

2.3 The ORM approach

In recent years, new frameworks and libraries have been developed

to provide better integration between the application and the DBMS.
Such frameworks can be separated into two categories. The first cate-
gory includes query language integrated libraries [Microsoft, b, Squeryl,
jOOQ] that provide stylized library calls for relational operations.
While developers still need to understand query concepts (such as se-
lections and joins), they no longer need to have knowledge about query
language syntax, and using such libraries does not incur the same secu-
rity issues as embedding raw strings in program code. In another cate-
gory are object-relational mapping (ORM) frameworks [JBoss, Cooper
et al., 2007, Microsoft, a, Django]. These frameworks go one step be-
yond language integrated libraries by giving developers the ability to
interact with the DBMS using the language of the application logic.
With ORMs, developers label certain classes as persistent—usually
through annotations or configuration files—and the framework auto-
matically manages all persistently stored objects for the application:
when the application needs to retrieve objects that are stored in the
Random documents with unrelated
content Scribd suggests to you:
suppresses revolt of 1047, 80-88;
besieges Brionne, 85, 86;
supports Henry I. against Geoffrey Martel, 90;
captures Domfront and Alençon, 91-95;
banishes William, count of Mortain, 97-99;
suppresses revolt of William Busac, 99, 100;
suppresses revolt of William of Arques, 101-105;
marries Matilda of Flanders, 105-110;
resists invasion of 1054, 110-117;
concludes peace with King Henry I., 116, 117;
founds Breteuil, 117;
engages in war of Ambrières, 118-120;
founds Roche-Mabille, 120;
defeats invasion of 1058, 120-123;
wins battle of Varaville, 122, 123;
receives commendation of Herbert II. of Maine, 118, 130;
invasion and conquest of Maine, 132-135;
takes possession of Le Mans, 133, 134;
illness of, 136, 137;
engages in Breton campaign of 1064, 137-139;
relief of Dol and siege of Dinan, 139, 140;
position of, at close of 1064, 140-142;
first visit to England, 146;
claim to English throne, 150-152;
receives oath from Harold, earl of Wessex, 153-157;
negotiations with foreign powers, 160;
submits his cause to the pope, 161-163;
gathers an army, 164, 165, 168;
and fleet, 165-168;
demands fulfilment of Harold’s oath, 170;
receives Earl Tostig, 172;
preparations for the invasion of England, 180, 181;
delayed at the Dive estuary, 182, 183;
at St. Valery, 183, 184;
voyage to England, 185;
lands at Pevensey, 185, 186;
builds castles at Pevensey and Hastings, 187, 188;
devastations in Sussex, 188;
receives a message from Harold, 189, 190;
battle of Hastings, 195-206;
generalship of, 197, 198;
moves out of Hastings to the English position, 199;
details of battle, 200-206;
causes Harold’s burial on the shore of Hastings, 207;
takes quarters at Hastings, 211;
march on London, 215-219;
burns Southwark, 219, 220;
crosses Thames at Wallingford, 220;
receives submission of English leaders, 221, 222;
receives offer of the crown, 222, 223;
dealings with “Esegar” the Staller, 224, 225;
coronation, 225, 226;
builds Tower of London, 227;
extent of his authority 1066–7, 230, 231;
at Barking, 227, 233, 234;
grants charter to citizens of London, 240, 241;
visits Normandy, 241, 243-246;
return, 246;
suppresses revolt of Exeter, 253-257;
progress in Cornwall, 259, 260;
at Westminster for Matilda’s coronation, 260;
northern campaign of 1068, 262-266;
receives submission of Malcolm III., 265, 266;
appoints Robert de Comines earl of Northumbria, 267;
second visit to York, 269;
in forest of Dean, 272;
march on Lindsey, 275, 276;
at Stafford, 276, 277;
at Nottingham, 278;
at Pontefract, 278, 279;
harrying of Northumbria, 279-281, 286, 287;
Christmas feast at York, 282;
march to the Tees and return to York, 282, 283;
march to Chester, 283-285;
agreement with Earl Asbiorn, 285;
campaign of Ely, 297-299;
relations with Robert the Frisian, 306, 307;
suppression of Mancel rising, 312, 313;
campaign of La Flèche, 313, 314;
concludes peace of Blanchelande, 314-317;
relations with Malcolm III., 317-323;
treaty of Abernethy, 323;
creation of earldom of Richmond, 324;
dealings with Edgar the Etheling, 325-327;
relations with, and condemnation of, Earl Waltheof, 336-340;
engages in Breton war of 1076, 341, 342;
last phase of reign, its character, 344, 345;
relations with Robert, 345, 346, 348-350;
campaign of Gerberoi, 347, 348;
movements during 1080, 351;
expedition into Wales, 353, 354;
arrest of Odo of Bayeux, 355-358;
death of Queen Matilda, 358, 359;
campaign of St. Suzanne, 359-361;
prepares for a Scandinavian invasion, 363, 364;
takes Oath of Salisbury, 364-366;
last departure from England, 366;
campaign of Mantes, 368, 369;
mortal injury of, 369;
illness of, 370-373;
disposition of inheritance, 371;
release of prisoners, 372;
death, 373;
burial, 374, 375;
ecclesiastical ideas of, 376-378, 381;
reform of English church, 388-402;
relations with the Curia, 402, 403;
 administrative changes introduced by,—see under castles,
Commune Concilium, Curia Regis, earldoms, fyrd, knight
service, private jurisdiction, sheriffs, writs;
orders taking of Domesday Inquest, 457
William II., king of England, 219, 233, 369, 371
William fitz Osbern, earl of Hereford, 93, 117, 230, 231, 261, 269,
287, 328, 414, 425-427, 433, 492
—— contributes to fleet, 166
—— regent of England, 243, 246, 415
—— death, 306
William Peverel, 451
William of Poitiers, biographer, 90, 381
William, archbishop of Rouen, 403, 404
William de Warenne, 243, 397
Wimund, commander of Moulins, 105
Winchelsea, 251
Winchester, 218, 219, 254, 260, 269, 277, 282, 337, 364, 419
Witanagemot, 17, 20
—— nature of, 410-414
Woodstock (Oxfordshire), 420
Worcestershire, 333, 426
Writs, the king’s, 168, 254, 420, 421
—— early Anglo-Norman, 227-231
Wulf, son of King Harold, 372
Wulfnoth, son of Earl Godwine, 212, 372
Wulfstan, bishop of Worcester, 230, 333, 391
Wulfwig, bishop of Dorchester, 232
Wyce, valley of, 354
 Y

York, 49, 171, 177, 190, 191, 252, 254, 262-265, 268, 269, 273-
275, 279, 282-285
—— church of, 335, 385
—— kingdom of, 7
 Footnotes

1. From the Catalogue of English Coins in the British Museum, Anglo-Saxon

Series.

2. From the Traité de Numismatique du Moyen Age, by Arthur Engel and

Raymond Serrure.

3. From the Handbook of the Coins of Great Britain and Ireland in British
Museum.

4. The boundary of the Danelaw in its full extent is proved by certain

twelfth-century lists of shires which divide England into “Westsexenelage,”
“Mirchenelage,” and “Danelage.” With regard to earlier times, the territory of
the Five Boroughs is delimited by the fiscal peculiarities described below
(Chapter XII.), and the kingdom of Northumbria substantially corresponds
with Yorkshire as surveyed in Domesday Book, but it is very uncertain how far
Guthrum’s kingdom extended westward after his final peace with Alfred.
London was annexed to Wessex, but the boundary does not seem to have
coincided in any way with the later county divisions.

5. See below, Chapter XII.

6. Chadwick, Studies in Anglo-Saxon Institutions, chapter v.

7. Chadwick, op. cit.

8. Maitland, Domesday Book and Beyond, 167.

9. See the account of the council at Bretford, below, page 61.

10. See Plummer, Life and Times of Alfred the Great, 67.

11. “Unready” here represents the A. S. unrædig—“devoid of counsel”—and

is applied to Ethelred because of his independence of the advice of the witan.

12. E. H. R., vii., 209.

13. See Eckel, Charles le Simple.

14. This identification cannot be considered certain. See Flodoard, ed. P.

Lauer.

15. The main features of Norman society in the eleventh century are
described in outline by Pollock and Maitland, History of English Law, i.,
chapter iii., on which the following sketch is founded.

16. The scanty evidence which exists on this matter is summarised by

Pollock and Maitland, H. E. L., chapter iii., and by Haskins, E. H. R., Oct.,
1907.

17. See on this matter F. Lot, Fidèles ou Vassaux.

18. See Histoire Général de France, Les Premiers Capetiens, p. 90; also
Sœhnée, Catalogue des Actes d’Henri Ier No. 38.

19. See Bohmer’s Kirche und Staat in England und in der Normandie, 20.

20. The fullest account of Cnut’s reign is given by Freeman. Norman

Conquest i., chapter vi. Freeman was disposed to underrate the value of
Scandinavian evidence, and hence considered Cnut’s reign almost exclusively
from the English standpoint.

21. See the lives of Earls Eric and Eglaf in the notes to the Crawford
Charters, No. xii.

22. P. and M., i., 20.

23. The most recent discussion in detail of this episode is that of Plummer,
Two Saxon Chronicles, ii. Freeman’s attempt to clear Godwine of complicity
was marked by a very arbitrary treatment of the contemporary authorities.

24. Heimskringla, trans. Morris and Magnusson, vol. iii., p. 10.

25. Op. cit., p. 181.

26. This is the duty of “hospitium,” exemption from which was frequently
granted in Anglo-Norman charters.

27. Swegen, Godwine’s eldest son, went on pilgrimage to Jerusalem, and

died on his way back.

28. See the map of the earldoms in 1066 given by Freeman, Norman
Conquest, ii.

29. In the next generation there was a tradition that Gospatric had been
murdered by Queen Edith on her brother’s behalf, Florence of Worcester,
1065.

30. Victoria History of Northamptonshire, i., 262–3.

31. In addition to the future Conqueror one other child was born to Robert
and Arlette—a daughter named Adeliz, who married Count Enguerrand of
Ponthieu; and after Robert’s death Arlette herself became the lawful wife of a
Norman knight named Herlwin of Conteville, whose two sons, Odo, bishop of
Bayeux, and Robert, count of Mortain, play a considerable part in the
succeeding history.

32. Ralf Glaber, iv., 6.

33. De la Borderie, Histoire de Bretagne, iii., 8–12.

34. Round, Calendar of Documents Preserved in France, 526.

35. This grant rests solely on the authority of Ordericus Vitalis, but it is
accepted by Flach, Les origines de l’ancienne France, 528–530.

36. The meeting place of this council is only recorded by William of

Malmesbury, Gesta Regum, ii., 285.

37. Ordericus Vitalis, iii., 431.

38. Among contemporaries who made the journey may be mentioned Count
Fulk Nerra of Anjou and Archbishop Ealdred of York.

39. Ordericus, ii., 369. Tutorem sui, Ducis.

40. Gesta Regum, ii., 285.

41. Gesta Regum, ii., 285. “Normannia fiscus regalis erat.” Henry of
Huntingdon, 189.

42. This is the opinion of Luchaire, Institutions monarchiques, ii., 17.

43. William of Jumièges, vii., 3.

44. Round, Calendar of Documents Preserved in France, No. 37.

45. Round, Calendar, No. 251.

46. Luchaire, Institutions monarchiques, ii., 233.

47. This is asserted very strongly by Freeman, ii., 201, and is implied by
Luchaire, Les Premières Capétiens, 163.

48. The whole story of the duke’s ride from Valognes to Falaise rests upon
the sole authority of Wace, and is only given here as a matter of tradition.

49. The topography of the battle is derived from Wace.

50. William of Poitiers, 81.

51. Ordericus Vitalis (iii., 342) makes a pointed reference to the length of
time occupied by the present siege in comparison with the capture of Brionne
in a single day by Robert of Normandy in 1090. But it is impossible to accept
his statement that the resistance of Guy of Burgundy was protracted for three
years.

52. William of Poitiers, 81: “Bella domestica apud nos in longum sopivit.”

53. In the imperfectly feudalised state of England a stricter doctrine seems

to have prevailed: see, on Waltheof’s case below, page 338.

54. This rests on no better authority than Wace. We know with more
certainty that the lands which Grimbald forfeited were bestowed by William
upon the See of Bayeux, of which Odo, the duke’s brother, became bishop in
1048.—Eng. Hist. Rev., xxii., 644.

55. “Vicissitudinem post hæc ipse Regi fide studiosissima reddidit.”

56. William of Poitiers, 82.

57. William of Poitiers, 82.

58. William of Poitiers, 87.

59. William of Poitiers, 88.

60. William of Jumièges, vii., 18.

61. William of Jumièges, vii., 18. The duke’s oath is given by Wace: Roman
de Rou, 9468.

62. William of Poitiers, 89.

63. William of Jumièges, vii., 19.

64. William of Jumièges, vii., 20.

65. The visit of William to England in 1051 will be considered below,

Chapter IV., in its bearing upon the general question of the English
succession.

66. William of Poitiers, 92.

67. This is definitely asserted by William of Malmesbury.

68. See on this episode, Round, Feudal England, 382–385.

69. Page 95.

70. William of Jumièges, vii., 7.

71. Labbè Concilia, xi., 1412.

72. For example, Freeman, N. C., iii., 92.

 73. Count Baldwin III. assumed the title of Marquis on the coins which he
issued.

74. Vita Eadwardi (R.S.), 404.

75. Page 97. On this question there is a conflict of evidence William of

Jumièges, whose authority is only second to that of William of Poitiers,
definitely asserts Geoffrey’s participation in the campaign. See Halphen, Conté
d’Anjou, 77. On the other hand, although the argument from the silence of
William of Poitiers should not be pressed too far, the terms of the treaty of
1053 (see below) certainly suggest that the king held Geoffrey guilty of a
breach of feudal duty, and later writers, such as Orderic, cannot be trusted
implicitly in regard to the detailed history of this period.

76. William of Poitiers, 99.

77. See note, page 112.

78. William of Jumièges, vii., 25.

79. See The Laws of Breteuil, by Miss M. Bateson, Eng. Hist. Rev., xx.

80. William of Poitiers, 99, 100.

81. In a charter abstracted by Round, Calendar of Documents Preserved in

France, No. 1256, there is a reference to a knight named Richard who was
seized by mortal illness while defending the frontier post of Châteauneuf-en-
Thimerais in this campaign.

82. William of Poitiers, 101. Wace gives topographical details.

83. William of Jumièges, vii., 28. The battle of Varaville led to the king’s
retreat, but a sporadic war lasted till 1060. It is probable that Norman
chroniclers have attached more importance to the battle than it really
possessed.

84. See Halphen, Comté d’Anjou, p. 133.

85. The history of Maine at this period has recently been discussed by
Flach, Les origines de l’ancienne France, vol. iii., p. 543–9.

86. The native Mancel authorities have little to say about the war of 1063,
the course of which is described by William of Poitiers, 103 et seq.

87. See the table on page 506.

88. Round. Calendar of Documents Preserved in France, No. 937.

89. Rhiwallon was brother of Junquené, the archbishop of Dol, whose

presence at the Norman court during William’s minority has been noted
above. De la Borderie, iii., p. [missing].

90. William of Poitiers (109–112) is the sole authority for this war and he
gives no dates. He definitely asserts the presence of Harold and his
companions in the Norman army, and his narrative contains nothing
irreconcilable with the relevant scenes in the Bayeux tapestry. The war was
probably intended to enforce Norman suzerainty over Brittany, and the rising
of Rhiwallon of Dol probably gave William his opportunity. De la Borderie,
Histoire de Bretagne, iii., p. [missing].

91. The canons of Chartres celebrated his obit on December 11th, a fact
which discounts the story in William of Jumièges that Conan was poisoned by
an adherent of William. If William had wished to remove Conan the latter
would certainly have died before William had sailed for England.

92. The scheme of policy which Green (Conquest of England, 522–524, ed.
1883) founded in relation to their marriage rests upon this assumption.

93. Poem in Worcester Chronicle, 1057.

94. Vita Eadwardi Confessoris (R. S.), 410.

95. Worcester Chronicle, 1042: “All the people chose Edward and received
him for King, as it belonged to him by right of birth.”

96. Chadwick, Studies in Anglo-Saxon Institutions, Excursus iv., p. 355.

97. The one contemporary account of Harold’s oath which we possess is

that given by William of Poitiers (ed. Giles, 108). According to this Harold
swore (1) to be William’s representative (vicarius) at Edward’s court; (2) to
work for William’s acceptance as king upon Edward’s death; (3) in the
meantime to cause Dover castle to receive a Norman garrison, and to build
other castles where the duke might command in his interest. In a later
passage William of Poitiers asserts that the duke wished to marry Harold to
one of his daughters. In all this there is nothing impossible, and to assume
with Freeman that the reception of a Norman garrison into a castle entrusted
to Harold’s charge would have been an act of treason is to read much later
political ideas into a transaction of the eleventh century. William was Edward’s
kinsman and we have no reason to suppose that the king would have
regarded with disfavour an act which would have given his cousin the means
of making good the claim to his succession which there is every reason to
believe that he himself had sanctioned twelve years before.

98. Vita Edwardi Confessoris (R. S.), 432.

99. William of Poitiers, 123.

100. William of Malmesbury, Gesta Regum, ii., 299.

 101. The statement that William promised, if successful, to hold England as
a fief of the papacy is made by no writer earlier than Wace, who has no
authority on a point of this kind.

102. Monumenta Gregoriana.

103. Round, Geoffrey de Mandeville, 8.

104. William of Poitiers, 124.

105. William of Malmesbury, Gesta Regum.

106. The list followed here is that printed by Giles as an appendix to the
Brevis Relatio. Scriptores, p. 21.

107. Guy of Amiens, 34: “Appulus et Caluber, Siculus quibus jacula fervet.”

108. Kingsley, Hereward the Wake, ed. 1889, p. 368.

109. This was Freeman’s final view. N. C., iii., 625.

110. Florence of Worcester, 1066.

111. Ordericus Vitalis, ii., 120.

112. Chronicles of Abingdon, Peterborough, and Worcester, 1066.

113. John of Oxenedes, a thirteenth-century monk of St. Benet of Holme,

asserts that Harold entrusted the defence of the coast to Ælfwold, abbot of
that house. The choice of an East Anglian abbot suggests that his
appointment was intended as a precaution against the Scandinavian danger.

114. See Introduction, above, page 48.

115. Heimskringla, page 165.

116. Simeon of Durham, 1066.

117. This episode forms the last entry in the Abingdon version of the
Chronicle, and it is described in a northern dialect.

118. Round, Calendar of Documents preserved in France, No. 1713.

119. William of Poitiers, 122.

120. W. P., 123. “Turmas militum cernens, non exhorrescens.”

121. Guy of Amiens, ed. Giles, 58.

122. William of Malmesbury, Gesta Regum, ii., 300.

 123. William of Poitiers, 125.

124. William of Poitiers, 126.

125. Abingdon Chronicle, 1066.

126. Guy of Amiens: “Diruta quae fuerant dudum castella reformas; Ponis
custodes ut tueantur ea.”

127. W. P.: “Normanni previa munitione Penevesellum, altera Hastingas

occupavere.”

128. See on this point Round, Feudal England, 150–152.

129. William of Poitiers, 128.

130. William’s real numbers probably lay between six and seven thousand.

131. See the paraphrase of this passage in the Roman de Rou, Freeman, N.
C., iii., 417.

132. Guy of Amiens, p. 31: “Ex Anglis unus, latitans sub rupe marina Cemit
ut effusas innumeras acies. Scandere currit equum; festinat dicere regi.”

133. Gaimar, l’Estoire des Engles, R. S., i., p. 222. Gaimar wrote in the
twelfth century, but he followed a lost copy of the A.-S. chronicle.

134. For the chronology of the campaigns of Stamfordbridge and Hastings

the dates given by Freeman are followed here.

135. Worcester Chronicle, 1066: “He com him togenes at thœre haran
apuldran.”

136. The statement that Harold further strengthened his position by

building a palisade in front of it rests solely on an obscure and probably
corrupt passage in the Roman de Rou (lines 7815 et seqq). Apart altogether
from the textual difficulty, the assertion of Wace is of no authority in view of
the silence both of contemporary writers and of those of the next generation.
In regard to none of the many earlier English fights of this century have we
any hint that the position of the army was strengthened in this manner; nor in
practice would it have been easy for Harold to collect sufficient timber to
protect a front of 800 yards on the barren down where he made his stand.
The negative evidence of the Bayeux tapestry is of particular importance
here; for its designer could represent defences of the kind suggested when he
so desired, as in the case of the fight at Dinan.

137. Spatz, p. 30, will only allow to William a total force of six to seven
thousand men.
 138. W. P., 133. “Cuncti pedites consistere densius conglobati.” For the
arrangement of the English army on the hill see Baring, E. H. R., xx., 65.

139. It is probable that the expressions in certain later authorities (e.g. W.

M., ii., 302, “pedites omnes cum bipennibus conserta ante se testudine”) from
which the formation by the English of a definite shield or wall has been
inferred mean no more than this. The “bord weal” of earlier Anglo-Saxon
warfare may also be explained as a poetical phrase for a line of troops in
close order.
See Round, Feudal England, 360–366.

140. This fact, which must condition any account to be given of the battle
of Hastings, was first stated by Dr. W. Spatz, “Die Schlacht von Hastings,”
section v., “Taktik beider Heere,” p. 34.

141. This point is brought out strongly by Oman, History of the Art of War.

142. Spatz, p. 29, uses this fact to limit the numbers of the Norman army.

143. W. P., 132.

144. Guy of Amiens: “Lævam Galli, dextram petiere Britanni. Dux cum
Normannis dimicat in medio.”

145. W. P., 132.

146. Florence of Worcester, 1066: “Ab hora tamen diei tertia usque ad
noctis crepusculum.”

147. Guy of Amiens. W. P., 133: “Cedit fere cuncta Ducis acies.”

148. “Fugientibus occurrit et obstitit, verberans aut minans hasta.”—W. P.,

134.

149. Bayeux tapestry scene: “Hic Odo episcopus, baculum tenens, confortat
pueros.”

150. W. P., 134.

151. “Animadvertentes Normanni ... non absque nimio sui incommodo

hostem tantum simul resistentem superari posse.”—W. P., 135.

152. “Normanni repente regirati equis interceptos et inclusos undique

mactaverunt.”—W. P., 135.

153. “Bis eo dolo simili eventu usi.”—William of Poitiers, 135.

154. “Languent Angli, et quasi reatum ipso defectu confitentes, vindictum

patiuntur.”—W. P., 135.
 155. Baring, E. H. R., xxii., 71.

156. “Jam inclinato die.”—W. P., 137. Crepusculi tempore.—Florence of

Worcester, 1066.

157. Baring, E. H. R., xxii., 69.

158. Guy of Amiens.

159. See the Waltham tract, De Inventione Sancti Crucis, ed. Stubbs.
William of Malmesbury was evidently acquainted with this legend.

160. It is probable that Wulfnoth had been taken together with Harold by
Guy of Ponthieu, and had been left behind in Normandy as a surety for the
observance of his brother’s oath to William.

161. Gesta Regum, R. S., 307.

162. Thomas Stubbs, ed. Raine; Historians of the Church of York, R. S., ii.,
100.

163. William of Poitiers, 139.

164. William of Poitiers, 139.

165. Guy of Amiens, 607.

166. William of Poitiers, 140.

167. Guy of Amiens, 617.

168. The embassy to Winchester is only mentioned by Guy of Amiens, who

omits all reference to William’s illness, which is derived from William of
Poitiers. Guy, however, places the message at this point of the campaign.

169. Round, Geoffrey de Mandeville, 4.

170. This is clearly meant by the statement of William of Poitiers that

William’s troops burned “quicquid ædificiorum citra flumen invenere.”

171. William of Poitiers, 141.

172. The Worcester Chronicle, followed by Florence of Worcester, 1066,

asserts that Edwin and Morcar submitted at “Beorcham,” but William of
Poitiers, whose authority is preferable on a point of this kind, implies that they
did not give in their allegiance until after the coronation. On the geography
relating to these events see Baring, E.H.R. xiii., 17.

173. William of Poitiers, 142.

174. Guy of Amiens, 687 et seqq.

 175. William of Poitiers, 143.

176. “Vehementer trementem,” Ordericus Vitalis, ii., 157.

177. Florence of Worcester, 1066.

178. William of Poitiers, 147–8.

179. This writ was issued in favour of one Regenbald, who had been King
Edward’s chancellor. It was printed by Round in Feudal England, 422, with
remarks on its historical importance.

180. Monasticon, i., 383. See also Round, Commune of London, 29.

181. Monasticon, i., 301. The date assigned here to these documents, of
which the text in the Monasticon edition is very faulty, is a matter of
inference; but the personal names which occur in them suggest that they
should be assigned to the very beginning of William’s reign.

182. Round, Calendar of Documents Preserved in France, No. 1423. See

also Commune of London, 30.

183. William of Poitiers, 148; Ordericus Vitalis, ii., 165.

184. Peterborough Chronicle, 1066. “And menn guldon him gyld ... and
sithan heora land bohtan.”—D. B., ii., 360. “Hanc Terram habet abbas ...
quando redimebant Anglici terras suas.” The combination of these statements
led Freeman to make the suggestion referred to in the text.

185. It may be noted that there exist a few proved cases in which a
Norman baron had married the daughter of his English predecessor, so that
here the king’s grant to the stranger would only confirm the latter in
possession of his wife’s inheritance.

186. D. B., i., 285 b. (Normanton on Trent).

187. Victoria History of Northamptonshire, i., 324.

188. Frequently printed, e.g., by Stubbs, Select Charters, 82.

189. Suggested by Round, Geoffrey de Mandeville, 439.

190. Ordericus Vitalis, ii., 167. The mercenaries were paid off at Pevensey
before William sailed for Normandy.

191. Peterborough Chronicle, 1087.

192. William of Poitiers (149) states that William Fitz Osbern was left in
charge of the city “Guenta,” which is described as being situated fourteen
miles from the sea which divides the English from the Danes, and as a point
where a Danish army might be likely to land. These indications imply that
Norwich (Venta Icenorum) was Fitz Osbern’s headquarters, although the
name Guenta alone would naturally refer to Winchester (Venta Belgarum).
The joint regency of Odo and William is asserted by Florence of Worcester,
1067, and the phrase in William of Poitiers, that Fitz Osbern “toto regno
Aquilionem versus præesset,” suggests that the Thames was the boundary
between his province and that of Odo. The priority of Fitz Osbern in the
regency is suggested by the fact that in a writ relating to land in Somerset, he
joins his name with that of the king in addressing the magnates of the shire.
Somersetshire certainly formed no part of his direct sphere of administration
at the time. For further references to this writ see below, Chapter XI.

193. The fullest list of names is given by Orderic, ii., 167.

194. William of Poitiers, 155.

195. Ordericus Vitalis, ii., 170.

196. Simeon of Durham, under the year 1072. He asserts that Oswulf
himself slew Copsige in the door of the church.

197. Simeon of Durham, under 1070.

198. Florence of Worcester, 1067.

199. Ordericus Vitalis, ii., 173.

200. The fullest account of the affair at Dover is given by Orderic (ii., 172–
5), who expands the slighter narrative of William of Poitiers.

201. Ordericus Vitalis, ii., 178.

202. Ordericus Vitalis., ii., 179.

203. “Ad Danos, vel alio, unde auxilium aliquod speratur, legatos
missitant.”—William of Poitiers, 157.

204. The story of the revolt of Exeter is critically discussed by Round,

Feudal England, 431–455.

205. Worcester Chronicle, 1067; Florence of Worcester, 1068; William of

Malmesbury, Gesta Regum, ii., 312.

206. The source of our information is an original charter granted by William

to the church of St. Martin’s le Grand on May 11th.—E. H. R. xii., 109.

207. Ordericus Vitalis, ii., 183.

208. The rising of Edwin and Morcar is not mentioned by the English
authorities, which are only concerned with the movements of Edgar and his
companions. Florence of Worcester says that the latter fled the court through
the fear of imprisonment. They had given no known cause of offence since
their original submission, but it is probable that they would have been kept in
close restraint if they had been in the king’s power when the northern revolt
broke out and that they fled to avoid this.

209. Ordericus Vitalis, ii., 184.

210. Ordericus Vitalis, ii., 185.

211. Simeon of Durham, 1069.

212. Ordericus Vitalis, ii., 188. From his statement that Earl William beat the
rebels “in a certain valley,” it is evident that the military operations were not
confined to the city of York.

213. Ordericus Vitalis, ii., 189.

214. For the events of 1069 Orderic is almost the sole authority, and his
narrative is not always easy to follow. On the other hand he is doubtless in
great part following the contemporary William of Poitiers, and his tale is quite
consistent with itself if due allowance is made for its geographical confusion.

215. The exact scene of Waltheof’s exploit is uncertain. Orderic implies that
the entire Norman garrison in York perished in the unsuccessful sally. Florence
of Worcester states that the castles were taken by storm. The latter is
certainly the more probable, and agrees better with the tradition, preserved
by William of Malmesbury, of the slaughter at the gate. The gate in question,
on this reading of the story, will belong to one of the castles; it cannot well be
taken to be one of the gates of the town.

216. The mutilation is only recorded by a late authority, the Winchester

Annals.

217. Ordericus’ narrative at this point is not very clear, but this is probably
his meaning.

218. By Ordericus William is made to return to York through Hexham

(“Hangustaldam revertabatur a Tesca”). This being impossible it is generally
assumed that Helmsley (Hamilac in D. B.) should be read for Hexham, in
which case William would probably cross the Cleveland hills by way of
Bilsdale.

219. “Desertores, vero, velut inertes, pavidosque et invalidos, si discedant,

parvi pendit.”

220. Chester castle was planted within arrow shot of the landing stage on
the right bank of the Dee, and also commanded the bridge which carried the
road from the Cheshire plain to the North Wales coast.
 221. Peterborough Chronicle, 1069.

222. William of Malmesbury, Gesta Pontificum, § 420.

223. Domesday Book, i., 346.

224. Peterborough Chronicle, 1070.

225. The passages which follow are founded on the narrative of Hugh
“Candidus,” a monk of Peterborough, who in the reign of Henry II. wrote an
account of the possessions of the abbey, and inserts a long passage
descriptive of the events of 1070. The beginning of his narrative agrees
closely with the contemporary account in the Peterborough Chronicle, but his
tale of the doings of the Danes in Ely after the sack of Peterborough is
independent, and bears every mark of truth. Wherever it is possible to test
Hugh’s work, in regard to other matters, its accuracy is confirmed. See Feudal
England, 163, V.C.H. Notts, i., 222. Hugh’s Chronicle has not been printed
since its edition by Sparke in the seventeenth century.

226. Ordericus Vitalis, ii., 216. The death of Edwin formed the conclusion of
the narrative of William of Poitiers as Orderic possessed it.

227. Florence of Worcester, 1070.

228. Historia Eliensis, 240.

229. Ordericus Vitalis, ii., 216.

230. Florence of Worcester, 1071.

231. Historia Eliensis, 245.

232. See “Ely and her Despoilers,” in Feudal England, 459.

233. Gaimar, L’estoire des Engles, R. S.

234. Gesta Herewardi, R. S.

235. See Varenbergh, Relations Diplomatiques entre le comté de Flandre et

l’Angleterre. Luchaire, Les Premiers Capetiens, 169.

236. Halphen, Comté d’Anjou, 180, has shown that Azo had appeared in
Maine by the spring of 1069.

237. The authorities for the present war are the history of Ordericus Vitalis
and the life of Bishop Arnold of Le Mans, ed. Mabillon; Vetera Analecta.

238. “Facta conspiratione quam communionem vocabant.”—Vet. An., 215.

239. Gesta Regum, ii., 316.

 240. Vetera Analecta, 286.

241. Hoel, unlike his predecessors, followed a policy of friendship towards

Anjou, and restored to Fulk le Rechin the conquests made by Count Conan on
the Angevin march. De la Borderie, iii., 26.

242. The terms of the peace of Blanchelande are given by Orderic.

243. E. H. R., xx., 61.

244. See table H.

245. Simeon of Durham, 1072.

246. This third flight of Edgar to Scotland rests solely upon the authority of
Simeon of Durham, and it is quite possible that the latter may have been
confused about the course of events at this point.

247. Worcester Chronicle, 1073.

248. Brian’s tenure of the earldom of Richmond is proved by a charter to

the priory of St. Martin de Lamballe, in which lands are granted by “Brientius,
comes Anglica terra.” (De la Borderie, iii., 25.) As Brian’s father, Count Éon of
Penthievre, did not die before 1079 the title “comes” cannot refer to any
French county possessed by Brian. As in the eleventh century every “earldom”
consisted of a shire or group of shires, it would seem to follow that
Richmondshire at this date was regarded as a territorial unit distinct from
Yorkshire.

249. Norman Conquest, iv., 517.

250. Worcester Chronicle, 1075.

251. Worcester Chronicle, 1075.

252. According to Wace Ralf had served among the Breton auxiliaries at the
battle of Hastings.

253. Ordericus Vitalis, ii., 258 et seq.

254. Florence of Worcester, 1074.

255. Worcester Chronicle, 1076.

256. Epistolæ Lanfranci.

257. Florence of Worcester, 1074.

258. It does not appear that any medieval historian regarded this as an act
of treachery on Waltheof’s part.
 259. F. N. C., iv., 585.

260. Gesta Regum, ii., 312.

261. This point is made by Pollock and Maitland. H. E. L., i., 291.

262. For the rest of the Conqueror’s reign, there was peace between
Normandy and Brittany, except that in 1086 William, to whom the new count
Alan Fergant, the son of Hoel, had refused homage, crossed the border once
more and laid siege to Dol. In this siege also he was unsuccessful, and
speedily came to terms with Alan, who received Constance, the Conqueror’s
daughter, in marriage.

263. Simeon of Durham, 1075.

264. Ordericus Vitalis, ii., 290.

265. Ordericus Vitalis, ii., 259.

266. Charter of King Philip to St. Quentin, Gallia Christ; X. Inst. 247. Among
the witnesses are Anselm of Bec, and Ives de Beaumont, the father-in-law of
Hugh de Grentemaisnil.

267. Worcester Chronicle, 1079.

268. S. D., Gesta Regum, 1080.

269. Round, Calendar, No. 1114.

270. Ibid., 1113.

271. Ibid., 78.

272. Orderic, ii., 315.

273. This fact is of importance, as giving an example, rare in England, of a

true “vicecomes,” an earl’s deputy as distinguished from a sheriff.

274. For all these events Simeon of Durham is the authority giving most
detail.

275. Hist. Monast. de Abingdon, ii., 10.

276. Brut y Tywysogion, 1080.

277. Mon. Angl., vii., 993, from an “inspeximus” of 31 Ed. I. The charter in
question is dated “apud villam Dontonam,” which in the index to the volume
of Patent Rolls is identified with Downton, Wilts. William, at Downton, may
very well have been on his way to one of the Hampshire or Dorset ports.
 278. iii., 168. On the other hand, Giesbrecht (iii., 531) has suggested that a
political difference was the occasion of the quarrel between Odo and William,
the former wishing to take up arms for Gregory VII., while the latter was on
friendly terms with the emperor. But Gregory himself in a letter addressed to
William (Register, viii., 60), while reproving his correspondent for lack of
respect towards his brother’s orders, admits that Odo had committed some
political offence against the king. As to the nature of that offence, we have no
contemporary statement, nor do we know how far Gregory may have
possessed accurate information as to the motives which induced William’s
action.

279. William of Malmesbury.

280. Ordericus Vitalis, iii., 196.

281. An isolated reference to the siege of Saint-Suzanne occurs in the

Domesday of Oxfordshire, in which county the manor of Ledhall had been
granted to Robert d’Oilly, “apud obsidionem S. Suzanne.”

282. Heimskringla, iii., 198.

283. The severity of the devastation should not be exaggerated, for in 1086
Lincolnshire, Norfolk, and Suffolk were the most prosperous parts of England.

284. Cnut’s preparations and death are described at length in his life by
Ethelnoth, printed in the Scriptores Rerum Danicarum.

285. Peterborough Chronicle, 1086.

286. See Flach, Les Origines de l’ancienne France, 531–534.

287. The ecclesiastical history of Normandy and England in the eleventh

century is treated by Böhmer, Kirche und Staat in England, und in der
Normandie, on which book this chapter is based.

288. See above, Introduction, ii., pp. 39, 40.

289. Especially in the Danelaw, V. C. H., Derby i., Leicester i.

290. Stubbs, Select Charters, 85. The writ in question probably belongs to
the year 1075.

291. Pollock and Maitland, i., 89.

292. Peterborough Chronicle, 1083.

293. Abbot Ethelhelm of Abingdon was considered to have offended in this

respect. Hist. Monast. de Abingdon, ii., 283.

294. See above, Chapter V.

 295. Ordericus Vitalis, ii., 315.

296. Easter, 1069: King William; Matilda, the Queen; Richard, the King’s
son; Stigand, archbishop of Canterbury; Ealdred, archbishop of York; William,
bishop of London; Ethelric, bishop of Selsey; Herman, bishop of Thetford;
Giso, bishop of Wells; Leofric, bishop of Exeter; Odo, bishop of Bayeux;
Geoffrey, bishop of Coutances; Baldwin, bishop of Evreux; Arnold, bishop of
Le Mans; Count Robert (of Mortain), Earl William Fitz Osbern, Count Robert of
Eu, Earl Ralf (of Norfolk?), Brian of Penthievre, Fulk de Alnou, Henry de
Ferrers; Hugh de Montfort, Richard the son of Count Gilbert, Roger d’ Ivri,
Hamon the Steward, Robert, Hamon’s brother.—Tardif, Archives de l’Empire,
179.
Christmas, 1077: King William; Lanfranc, archbishop of Canterbury;
Thomas, archbishop of York; Odo, bishop of Bayeux; Hugh, bishop of London;
Walkelin, bishop of Winchester; Remi, bishop of Lincoln; Maurice, the
chancellor; Vitalis, abbot of Westminster; Scotland, abbot of Ch. Ch.,
Canterbury; Baldwin, abbot of St. Edmunds; Simeon, abbot of Ely; Aelfwine,
abbot of Ramsey; Serlo, abbot of Gloucester; Earl Roger of Montgomery, Earl
Hugh of Chester, Count Robert of Mortain, Count Alan of Richmond, Earl
Aubrey of Northumbria, Hugh de Montfort, Henry de Ferrers, Walter Giffard,
Robert d’ Oilli, Hamon the Steward, Wulfstan, bishop of Worcester.—Ramsey
Chartulary, R. S., ii., 91.
Easter, 1080: King William; Matilda the Queen; Robert, the king’s son;
William, the king’s son; William, archbishop of Rouen; Richard, archbishop of
Bourges; Warmund, archbishop of Vienne; Geoffrey, bishop of Coutances;
Gilbert, bishop of Lisieux; Count Robert, the king’s brother; Count Roger of
Eu, Count Guy of Ponthieu, Roger de Beaumont, Robert and Henry, his sons,
Roger de Montgomery, Walter Giffard, William d’ Arques.—Calendar of
Documents Preserved in France, ed. J. H. Round, No. 78.

297. Printed in Transactions of Somerset Archæological and Historical

Society, xxiii., 56

298. Bath Chartulary (Somerset Record Society), i., 36.

299. Hist. Monasterii de Abingdon, R. S., ii., 9.

300. Ibid., 10.

301. Round, Calendar of Documents Preserved in France, No. 712.

302. Henry I. is seldom found north of Nottingham.

303. Monasticon, iii., 377.

304. V. C. H., Warwick, i., 258.

305. See above, Chapter VI.

 306. See the complaints of his aggressions in Heming’s History of the
Church of Worcester; Monasticon, i., 593–599.

307. William of Malmesbury, ii., 314.

308. Calendar of Documents Preserved in France, No. 77.

309. Ordericus Vitalis, ii., 178.

310. Compare Round, Geoffrey de Mandeville, 322.

311. See the charters of William II. in Monasticon, viii., 1167.

312. Ordericus Vitalis, ii., 219.

313. Reproduced herewith.

314. Wharton, Anglia Sacra, i., 339.

315. Maitland, Domesday Book and Beyond, 80-83.

316. Charter of William I., Monasticon, i., 477.

317. Foundation charter of Blyth Priory, Monasticon, iv., 623.

318. There is some evidence to suggest that the lord of a vill could cause a
court to be held there by his steward. This, however, is the result of
seignorial, not communal, ideas.

319. Round, Feudal England, 225–314, has given the clearest account of
the introduction and development of knight service in England.

320. Feudal England, as quoted above, page 447. See also Morris, Welsh
Wars of Edward, i., 36, arguing for a total of 5000.

321. Frequently printed, e.g. by Stubbs, Select Charters, 86.

322. Birch, Cartularium, i., 414.

323. Birch, Cartularium, iii., 671; Maitland, Domesday Book, 502.

324. Birch, Cartularium, iii., 671; Maitland, Domesday Book, 456.

325. Maitland, D. B. and Beyond, 4.

326. The fact that the assessment of southern and western England was
based upon a conventional unit of five hides was first enunciated by Mr. J. H.
Round in Feudal England.

327. Vinogradoff, E. H. R., xix., 282.

 328. Feudal England, 98–103.

329. For the “six-carucate unit” see Feudal England, 69. Victoria Histories,
Derby, Notts, Leicester, and Lincoln.

330. Feudal England, 42.

331. V. C. H., Derby, i., 295.

332. This was the view of Professor Maitland, Domesday Book and Beyond,
24.

333. The contemporary description of the Domesday Survey published by

Stevenson, E. H. R., xxii., 72, makes it probable that the bordars were in
theory distinguished from other classes by the fact that they possessed no
share in the arable fields of the vill.

334. See V. C. H., Hertford, i., 293.

335. V. C. H., Bedford, i., 200.

336. The former view is that of Mr. Round, the latter that of Professor
Maitland.

337. We also know that the returns were checked in each county by a
second set of commissioners who were deliberately sent by the king into
shires where they possessed no personal interest.—E. H. R., xxii., 72.

338. Feudal England, 141.

339. Dialogus de Saccario (ed. 1902), p. 108.

 Advertisements

A Selection from the

Catalogue of

G. P. PUTNAM’S SONS
Complete Catalogues sent
on application

Heroes of the Nations.

A series of biographical studies of the lives and work of a number of

representative historical characters about whom have gathered the great
traditions of the Nations to which they belonged, and who have been
accepted, in many instances, as types of the several National ideals.
With the life of each typical character will be presented a picture of the
National conditions surrounding him during his career.
The narratives are the work of writers who are recognized authorities
on their several subjects, and, while thoroughly trustworthy as history,
will present picturesque and dramatic “stories” of the Men and of the
events connected with them.
To the Life of each “Hero” will be given one duodecimo volume,
handsomely printed in large type, provided with maps and adequately
illustrated according to the special requirements of the several subjects.

Nos. 1–32, each $1.50

Half Leather 1.75
No. 33 and following Nos., each (By mail, $1.50, net
1.35)
Half Leather (by mail, $1.75) net 1.60
 For full list of volumes see next page.

HEROES OF THE NATIONS

NELSON. By W. Clark Russell.
GUSTAVUS ADOLPHUS. By C. R. L. Fletcher.
PERICLES. By Evelyn Abbott.
THEODORIC THE GOTH. By Thomas Hodgkin.
SIR PHILIP SIDNEY. By H. R. Fox-Bourne.
JULIUS CÆSAR. By W. Warde Fowler.
WYCLIF. By Lewis Sergeant.
NAPOLEON. By W. O’Connor Morris.
HENRY OF NAVARRE. By P. F. Willert.
CICERO. By J. L. Strachan-Davidson.
ABRAHAM LINCOLN. By Noah Brooks.
PRINCE HENRY (OF PORTUGAL) THE NAVIGATOR. By C. R. Beazley.
JULIAN THE PHILOSOPHER. By Alice Gardner.
LOUIS XIV. By Arthur Hassall.
CHARLES XII. By R. Nisbet Bain.
LORENZO DE’ MEDICI. By Edward Armstrong.
JEANNE D’ARC. By Mrs. Oliphant.
CHRISTOPHER COLUMBUS. By Washington Irving.
ROBERT THE BRUCE. By Sir Herbert Maxwell.
HANNIBAL. By W. O’Connor Morris.
ULYSSES S. GRANT. By William Conant Church.
ROBERT E. LEE. By Henry Alexander White.
THE CID CAMPEADOR. By H. Butler Clarke.
SALADIN. By Stanley Lane-Poole.
BISMARCK. By J. W. Headlam.
ALEXANDER THE GREAT. By Benjamin I. Wheeler.
CHARLEMAGNE. By H. W. C. Davis.
OLIVER CROMWELL. By Charles Firth.
RICHELIEU. By James B. Perkins.
DANIEL O’CONNELL. By Robert Dunlop.
SAINT LOUIS (Louis IX. of France). By Frederick Perry.
LORD CHATHAM. By Walford Davis Green.
OWEN GLYNDWR. By Arthur G. Bradley.
HENRY V. By Charles L. Kingsford.
EDWARD I. By Edward Jenks.
AUGUSTUS CÆSAR. By J. B. Firth.
FREDERICK THE GREAT. By W. F. Reddaway.
WELLINGTON. By W. O’Connor Morris.
CONSTANTINE THE GREAT. By J. B. Firth.
MOHAMMED. By D. S. Margoliouth.
CHARLES THE BOLD. By Ruth Putnam.
WASHINGTON. By J. A. Harrison.
WILLIAM THE CONQUEROR. By F. M. Stenton.

Other volumes in preparation are:

MOLTKE. By Spencer Wilkinson.

JUDAS MACCABÆUS. By Israel Abrahams.
SOBIESKI. By F. A. Pollard.
ALFRED THE TRUTHTELLER. By Frederick Perry.
FREDERICK II. By A. L. Smith.
MARLBOROUGH. By C. W. C. Oman.
RICHARD THE LION-HEARTED. By T. A. Archer.
WILLIAM THE SILENT. By Ruth Putnam.
GREGORY VII. By F. Urquhart.

New York—G. P. PUTNAM’S SONS, Publishers—London

The Story of the Nations.

 In the story form the current of each National life is distinctly
indicated, and its picturesque and noteworthy periods and episodes are
presented for the reader in their philosophical relation to each other as
well as to universal history.
It is the plan of the writers of the different volumes to enter into the
real life of the peoples, and to bring them before the reader as they
actually lived, labored, and struggled—as they studied and wrote, and as
they amused themselves. In carrying out this plan, the myths, with
which the history of all lands begins, will not be overlooked, though
these will be carefully distinguished from the actual history, so far as the
labors of the accepted historical authorities have resulted in definite
conclusions.
The subjects of the different volumes have been planned to cover
connecting and, as far as possible, consecutive epochs or periods, so
that the set when completed will present in a comprehensive narrative
the chief events in the great Story of the Nations; but it is, of course, not
always practicable to issue the several volumes in their chronological
order.
12o Illustrated, cloth, each $1.50
Half leather, each 1.75

For list of volumes see next page

THE STORY OF THE NATIONS

GREECE. Prof. Jas. A. Harrison.
ROME. Arthur Gilman.
THE JEWS. Prof. James K. Hosmer.
CHALDEA. Z. A. Ragozin.
GERMANY. S. Baring-Gould.
NORWAY. Hjalmar H. Boyesen.
SPAIN. Rev. E. E. and Susan Hale.
HUNGARY. Prof. A. Vámbéry.
CARTHAGE. Prof. Alfred J. Church.
THE SARACENS. Arthur Gilman.
THE MOORS IN SPAIN. Stanley Lane-Poole.
THE NORMANS. Sarah Orne Jewett.
PERSIA. S. G. W. Benjamin.
ANCIENT EGYPT. Prof. Geo. Rawlinson.
ALEXANDER’S EMPIRE. Prof. J. P. Mahaffy.
ASSYRIA. Z. A. Ragozin.
THE GOTHS. Henry Bradley.
IRELAND. Hon. Emily Lawless.
TURKEY. Stanley Lane-Poole.
MEDIA, BABYLON, AND PERSIA. Z. A. Ragozin.
MEDIÆVAL FRANCE. Prof. Gustave Masson.
HOLLAND. Prof. J. Thorold Rogers.
MEXICO. Susan Hale.
PHŒNICIA. George Rawlinson.
THE HANSA TOWNS. Helen Zimmern.
EARLY BRITAIN. Prof. Alfred J. Church.
THE BARBARY CORSAIRS. Stanley Lane-Poole.
RUSSIA. W. R. Morfill.
THE JEWS UNDER ROME. W. D. Morrison.
SCOTLAND. John Mackintosh.
SWITZERLAND. R. Stead and Mrs. A. Hug.
PORTUGAL. H. Morse-Stephens.
THE BYZANTINE EMPIRE. C. W. C. Oman.
SICILY. E. A. Freeman.
THE TUSCAN REPUBLICS. Bella Duffy.
POLAND. W. R. Morfill.
PARTHIA. Geo. Rawlinson.
JAPAN. David Murray.
THE CHRISTIAN RECOVERY OF SPAIN. H. E. Watts.
AUSTRALASIA. Greville Tregarthen.
SOUTHERN AFRICA. Geo. M. Theal.
VENICE. Alethea Wiel.
THE CRUSADES. T. S. Archer and C. L. Kingsford.
VEDIC INDIA. Z. A. Ragozin.
BOHEMIA. C. E. Maurice.
CANADA. J. G. Bourinot.
THE BALKAN STATES. William Miller.
BRITISH RULE IN INDIA. R. W. Frazer.
MODERN FRANCE. André Le Bon.
THE BRITISH EMPIRE. Alfred T. Story. Two vols.
THE FRANKS. Lewis Sergeant.
THE WEST INDIES. Amos K. Fiske.
THE PEOPLE OF ENGLAND. Justin McCarthy, M.P. Two vols.
AUSTRIA. Sidney Whitman.
CHINA. Robt. K. Douglass.
MODERN SPAIN. Major Martin A. S. Hume.
MODERN ITALY. Pietro Orsi.
THE THIRTEEN COLONIES. Helen A. Smith. Two vols.
WALES AND CORNWALL. Owen M. Edwards.
MEDIÆVAL ROME. Wm. Miller.
THE PAPAL MONARCHY. Wm. Barry.
MEDIÆVAL INDIA. Stanley Lane-Poole.
BUDDHIST INDIA. T. W. Rhys-Davids.
THE SOUTH AMERICAN REPUBLICS. Thomas C. Dawson. Two vols.
PARLIAMENTARY ENGLAND. Edward Jenks.
MEDIÆVAL ENGLAND. Mary Bateson.
THE UNITED STATES. Edward Earle Sparks. Two vols.
ENGLAND. THE COMING OF PARLIAMENT. L. Cecil Jane.
GREECE: EARLIEST TIMES TO A.D. 14. E. S. Shuckburgh.
ROMAN EMPIRE, THE, B.C. 29-A.D. 476. By N. Stuart Jones, M.A.
 Transcriptions of Genealogy Tables
Transcription of Table A
THE DUCAL HOUSE OF NORMANDY TABLE A

1 ‘Rollo’
11 William I.
111 Richard I.
1111 Richard II. m. Judith of Brittany
11111 Richard III.
111111 Nicholas, abbot of St Ouen
11112 Robert I.
111121 William II.
1111211 Robert II.
111122 Adeliz → (Table C)
11113 Adeliz m. Reginal, count of Burgundian Palatinate
111131 Guy of Brionne
11114 Malger, archbishop of Rouen
11115 William, count of Arques
1112 Robert, archbishop of Rouen
11121 Ralf de Wacy
1113 Malger count of Mortain
11131 William ‘the Warling’
1114 William, count of Eu
11141 Robert, count of Eu
11142 William, ‘Busac’
11143 Hugh, bishop of Lisieux
1115 Mahaut = Odo II of Blois
1116 Emma = Ethelred II.
11161 Alfred the Etheling
11162 Edward the Confessor
1117 Hawise = Geoffrey I. of Brittany →(Table B)

Transcription of Table B
The Counts of Brittany TABLE B

1 Geoffrey I. †1008 = Hawise of Normandy

11 Alan III. †1040
111 Conan II. o. s. p. 1066
112 Geoffrey ‘Grenonat’ (claimant 1075)
113 Hawise = Hoel, count of Cornouaille †1084
1131 Alan Fergent = Constance of Normandy †1119
12 Éon, count of Penthièvre. †1079
121 Brian, earl of Richmond
122 Alan, earl of Richmond

Transcription of Table C
The Descendents of Arlette TABLE C

1 Arlette
m. Robert of Normandy
11 William
12 Adeliz m. Enguerrand, count of Ponthieu
121 Adeliz m. Eudes, count of Champagne
1211 Stephen, count of Aumâl (claimant for England, 1095)
122 Judith m. Waltheof, earl of Northampton
1221 Maud m David I. king of Scots
m. Herlwin de Conteville
13 Odo, bishop of Bayeux
14 Robert, count of Mortain
15 Muriel

Transcription of Table D
Welcome to our website – the perfect destination for book lovers and
knowledge seekers. We believe that every book holds a new world,
offering opportunities for learning, discovery, and personal growth.
That’s why we are dedicated to bringing you a diverse collection of
books, ranging from classic literature and specialized publications to
self-development guides and children's books.

More than just a book-buying platform, we strive to be a bridge

connecting you with timeless cultural and intellectual values. With an
elegant, user-friendly interface and a smart search system, you can
quickly find the books that best suit your interests. Additionally,
our special promotions and home delivery services help you save time
and fully enjoy the joy of reading.

Join us on a journey of knowledge exploration, passion nurturing, and

personal growth every day!

ebookbell.com