Professional Documents
Culture Documents
Download textbook Automated Reasoning 7Th International Joint Conference Ijcar 2014 Held As Part Of The Vienna Summer Of Logic Vsl 2014 Vienna Austria July 19 22 2014 Proceedings 1St Edition Stephane Demri ebook all chapter pdf
Download textbook Automated Reasoning 7Th International Joint Conference Ijcar 2014 Held As Part Of The Vienna Summer Of Logic Vsl 2014 Vienna Austria July 19 22 2014 Proceedings 1St Edition Stephane Demri ebook all chapter pdf
https://textbookfull.com/product/automated-reasoning-10th-
international-joint-conference-ijcar-2020-paris-france-
july-1-4-2020-proceedings-part-ii-nicolas-peltier/
https://textbookfull.com/product/automated-reasoning-10th-
international-joint-conference-ijcar-2020-paris-france-
july-1-4-2020-proceedings-part-i-nicolas-peltier/
Learning and Collaboration Technologies Designing and
Developing Novel Learning Experiences First
International Conference LCT 2014 Held as Part of HCI
International 2014 Heraklion Crete Greece June 22 27
2014 Proceedings Part I 1st Edition Panayiotis Zaphiris
https://textbookfull.com/product/learning-and-collaboration-
technologies-designing-and-developing-novel-learning-experiences-
first-international-conference-lct-2014-held-as-part-of-hci-
international-2014-heraklion-crete-greece-june-2/
Automated Reasoning
7th International Joint Conference, IJCAR 2014
Held as Part of the Vienna Summer of Logic, VSL 2014
Vienna, Austria, July 19–22, 2014, Proceedings
123
Lecture Notes in Artificial Intelligence 8562
Automated Reasoning
7th International Joint Conference, IJCAR 2014
Held as Part of the Vienna Summer of Logic, VSL 2014
Vienna, Austria, July 19-22, 2014
Proceedings
13
Volume Editors
Stéphane Demri
New York University
Courant Institute of Mathematical Sciences
250 Mercer Street, New York, NY 10012, USA
E-mail: demri@lsv.ens.cachan.fr
Deepak Kapur
University of New Mexico
Department of Computer Science
Albuquerque, NM 87131-0001, USA
E-mail: kapur@cs.unm.edu
Christoph Weidenbach
Max Planck Institute for Informatics
Campus E1 4, 66123 Saarbrücken, Germany
E-mail: weidenbach@mpi-inf.mpg.de
In the summer of 2014, Vienna hosted the largest scientific conference in the
history of logic. The Vienna Summer of Logic (VSL, http://vsl2014.at) con-
sisted of twelve large conferences and 82 workshops, attracting more than 2000
researchers from all over the world. This unique event was organized by the Kurt
Gödel Society and took place at Vienna University of Technology during July
9 to 24, 2014, under the auspices of the Federal President of the Republic of
Austria, Dr. Heinz Fischer.
The conferences and workshops dealt with the main theme, logic, from three
important angles: logic in computer science, mathematical logic, and logic in
artificial intelligence. They naturally gave rise to respective streams gathering
the following meetings:
Mathematical Logic
The VSL keynote talks which were directed to all participants were given by
Franz Baader (Technische Universität Dresden), Edmund Clarke (Carnegie Mel-
lon University), Christos Papadimitriou (University of California, Berkeley) and
Alex Wilkie (University of Manchester); Dana Scott (Carnegie Mellon Univer-
sity) spoke in the opening session. Since the Vienna Summer of Logic contained
more than a hundred invited talks, it would not be feasible to list them here.
The program of the Vienna Summer of Logic was very rich, including not only
scientific talks, poster sessions and panels, but also two distinctive events. One
was the award ceremony of the Kurt Gödel Research Prize Fellowship Competi-
tion, in which the Kurt Gödel Society awarded three research fellowship prizes
endowed with 100.000 Euro each to the winners. This was the third edition of
the competition, themed Logical Mind: Connecting Foundations and Technology
this year.
The 1st FLoC Olympic Games formed the other distinctive event and were
hosted by the Federated Logic Conference (FLoC) 2014. Intended as a new FLoC
element, the Games brought together 12 established logic solver competitions
by different research communities. In addition to the competitions, the Olympic
Games facilitated the exchange of expertise between communities, and increased
the visibility and impact of state-of-the-art solver technology. The winners in
the competition categories were honored with Kurt Gödel medals at the FLoC
Olympic Games award ceremonies.
Organizing an event like the Vienna Summer of Logic was a challenge. We
are indebted to numerous people whose enormous efforts were essential in mak-
ing this vision become reality. With so many colleagues and friends working
with us, we are unable to list them individually here. Nevertheless, as rep-
resentatives of the three streams of VSL, we would like to particularly ex-
press our gratitude to all people who helped to make this event a success:
the sponsors and the Honorary Committee; the Organization Committee and
Foreword IX
the local organizers; the conference and workshop chairs and Program Commit-
tee members; the reviewers and authors; and of course all speakers and partici-
pants of the many conferences, workshops and competitions.
The Vienna Summer of Logic continues a great legacy of scientific thought
that started in Ancient Greece and flourished in the city of Gödel, Wittgenstein
and the Vienna Circle. The heroes of our intellectual past shaped the scientific
world-view and changed our understanding of science. Owing to their achieve-
ments, logic has permeated a wide range of disciplines, including computer sci-
ence, mathematics, artificial intelligence, philosophy, linguistics, and many more.
Logic is everywhere – or in the language of Aristotle,
This volume contains the papers presented at IJCAR’14: 7th International Joint
Conference on Automated Reasoning (IJCAR) held on July 19-22, 2014 in Vi-
enna. This year’s meeting was a merging of three leading events in automated
reasoning – CADE (International Conference on Automated Deduction), FroCoS
(International Symposium on Frontiers of Combining Systems) and TABLEAUX
(International Conference on Automated Reasoning with Analytic Tableaux and
Related Methods). IJCAR is the premier international joint conference on all
topics in automated reasoning, including foundations, implementations, and ap-
plications. Previous IJCAR conferences were held at Siena (Italy) in 2001, Cork
(Ireland) in 2004, Seattle (USA) in 2006, Sydney (Australia) in 2008, Edinburgh
(UK) in 2010 and Manchester (UK) in 2012.
IJCAR 2014 is part of Federated Logic Conference (FLoC) that is itself part
of Vienna Summer in Logic (VSL) and 24 workshops are affiliated with IJCAR.
The Vienna Summer of Logic is a unique event organized by the Kurt Gödel
Society at Vienna University of Technology from July 9 to 24, 2014.
The call for papers for IJCAR’14 invited authors to submit full papers (of
15 pages) and system descriptions (of 7 pages). There were 83 submissions (63
regular papers and 20 system descriptions) of which 37 were accepted (26 regular
papers and 11 system descriptions). Each submission was assigned to at least
three Program Committee members, who carefully reviewed the papers, with the
help of 116 external referees. We wish to thank the Program Committee members
and all their reviewers for their works and efforts in evaluating the submissions. It
was a pleasure to work with all of them. The EasyChair conference management
system was a great help in dealing with all aspects of putting our program and
the proceedings together.
IJCAR 2014 had invited talks by Rajeev Goré (The Australian National Uni-
versity) and Ken McMillan (Microsoft Research). In addition, IJCAR together
with other FLoC conferences, had two invited plenary talks by Véronique Cortier
(Centre National de la Recherche Scientifique) and by Orna Kupferman (Hebrew
University). These proceedings contain three papers and an abstract of these in-
vited talks. We want to thank the invited speakers for contributing to the success
of the IJCAR 2014.
Many people helped to make IJCAR 2014 a success. We want to thank the
the conference co-chairs and the organizing committee consisting of Christian
Fermüller, Stefan Hetzl and Giselle Reis, the publicity chair Morgan Deters and
the workshop chair Matthias Horbach. We are also indebted to the FLoC and
VSL organization committees.
XII Preface
Most importantly, we would like to thank all the authors for submitting their
work to IJCAR 2014: we believe the outcome is an exciting technical program.
Program Committee
Franz Baader TU Dresden, Germany
Peter Baumgartner National ICT Australia
Bernhard Beckert Karlsruhe Institute of Technology, Germany
Jasmin Christian
Blanchette TU München, Germany
Bernard Boigelot University of Liège, Belgium
Maria Paola Bonacina Universita‘ degli Studi di Verona, Italy
Agata Ciabattoni TU Wien, Austria
Koen Claessen Chalmers University of Technology, Sweden
Leonardo De Moura Microsoft Research, USA
Stéphanie Delaune CNRS, LSV, France
Stéphane Demri CNRS, France and NYU, USA
Stephan Falke Karlsruhe Institute of Technology (KIT),
Germany
Christian Fermüller TU Wien, Austria
Pascal Fontaine Loria, INRIA, University of Nancy, France
Silvio Ghilardi Università degli Studi di Milano, Italy
Jürgen Giesl RWTH Aachen, Germany
Valentin Goranko Technical University of Denmark
Radu Iosif Verimag/CNRS/University of Grenoble, France
Deepak Kapur University of New Mexico, USA
Boris Konev The University of Liverpool, UK
Konstantin Korovin Manchester University, UK
Daniel Kroening Oxford University, UK
Viktor Kuncak EPFL, Switzerland
Martin Lange University of Kassel, Germany
Stephan Merz Inria Lorraine, France
Aart Middeldorp University of Innsbruck, Austria
Enric Rodrı́guez Carbonell Technical University of Catalonia, Spain
Renate A. Schmidt University of Manchester, UK
Carsten Schuermann IT University of Copenhagen, Denmark
Roberto Sebastiani DISI, University of Trento, Italy
Viorica
Sofronie-Stokkermans University Koblenz-Landau, Germany
XIV Organization
Additional Reviewers
Alama, Jesse Gimenez, Stéphane
Areces, Carlos Gladisch, Christoph
Armas, Ana Graham-Lengrand, Stéphane
Artale, Alessandro Grebing, Sarah
Atkey, Robert Greco, Giuseppe
Audemard, Gilles Griggio, Alberto
Badban, Bahareh Herda, Mihai
Baldi, Paolo Heule, Marijn
Barrett, Clark Hoder, Krystof
Bellodi, Elena Horbach, Matthias
Bengtson, Jesper Hou, Zhe
Benzmüller, Christoph Huang, Guan-Shieng
Bezhanishvili, Nick Hustadt, Ullrich
Bormer, Thorsten Jacobs, Swen
Bresolin, Davide Jovanović, Dejan
Brock-Nannestad, Taus Kapur, Deepak
Brockschmidt, Marc King, Timothy
Bruns, Daniel Koopmann, Patrick
Bruse, Florian Kop, Cynthia
Bucheli, Samuel Kuraj, Ivan
Chen, Hong-Yi Lammich, Peter
Conchon, Sylvain Leitsch, Alexander
Cyriac, Aiswarya Lellmann, Bjoern
De Nivelle, Hans Lisitsa, Alexei
Della Monica, Dario Liu, Jun
Demri, Stephane Liu, Wanwei
Dietl, Werner Lozes, Etienne
Dyckhoff, Roy Ludwig, Michel
Eades Iii, Harley Lutz, Carsten
Ehlers, Rüdiger Madhavan, Ravichandhran
Enea, Constantin Marchi, Jerusa
Erbatur, Serdar Mccabe-Dansted, John
Ferreira, Francisco Mclaughlin, Sean
Fiorino, Guido Metcalfe, George
Franconi, Enrico Momigliano, Alberto
Galmiche, Didier Nagele, Julian
Organization XV
Véronique Cortier
The research leading to these results has received funding from the European
Research Council under the European Union’s Seventh Framework Programme
(FP7/2007-2013)/ERC grant agreement no 258865, project ProSecure.
And-Or Tableaux for Fixpoint Logics
with Converse: LTL, CTL, PDL and CPDL
Rajeev Goré
Abstract. Over the last forty years, computer scientists have invented
or borrowed numerous logics for reasoning about digital systems. Here,
I would like to concentrate on three of them: Linear Time Temporal
Logic (LTL), branching time Computation Tree temporal Logic (CTL),
and Propositional Dynamic Logic (PDL), with and without converse.
More specifically, I would like to present results and techniques on how
to solve the satisfiability problem in these logics, with global assump-
tions, using the tableau method. The issues that arise are the typical
tensions between computational complexity, practicality and scalability.
This is joint work with Linh Anh Nguyen, Pietro Abate, Linda Postniece,
Florian Widmann and Jimmy Thomson.
Structured Search and Learning
Kenneth L. McMillan
Microsoft Research
References
1. Bradley, A.R.: SAT-based model checking without unrolling. In: Jhala, R., Schmidt,
D. (eds.) VMCAI 2011. LNCS, vol. 6538, pp. 70–87. Springer, Heidelberg (2011)
2. McMillan, K.L.: Lazy annotation for program testing and verification. In: Touili,
T., Cook, B., Jackson, P. (eds.) CAV 2010. LNCS, vol. 6174, pp. 104–118. Springer,
Heidelberg (2010)
Table of Contents
Invited Papers
From Reachability to Temporal Specifications in Cost-Sharing
Games . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
Guy Avni, Orna Kupferman, and Tami Tamir
Electronic Voting: How Logic Can Help . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
Véronique Cortier
And-Or Tableaux for Fixpoint Logics with Converse: LTL, CTL,
PDL and CPDL . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26
Rajeev Goré
HOL
Unified Classical Logic Completeness: A Coinductive Pearl . . . . . . . . . . . . 46
Jasmin Christian Blanchette, Andrei Popescu, and Dmitriy Traytel
A Focused Sequent Calculus for Higher-Order Logic . . . . . . . . . . . . . . . . . . 61
Fredrik Lindblad
SMT
A Gentle Non-disjoint Combination of Satisfiability Procedures . . . . . . . . 122
Paula Chocron, Pascal Fontaine, and Christophe Ringeissen
Equational Reasoning
A Rewriting Strategy to Generate Prime Implicates in Equational
Logic . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 137
Mnacho Echenim, Nicolas Peltier, and Sophie Tourret
XXVI Table of Contents
Verification
Locality Transfer: From Constrained Axiomatizations to Reachability
Predicates . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 192
Matthias Horbach and Viorica Sofronie-Stokkermans
Proving Termination and Memory Safety for Programs with Pointer
Arithmetic . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 208
Thomas Ströder, Jürgen Giesl, Marc Brockschmidt, Florian Frohn,
Carsten Fuhs, Jera Hensel, and Peter Schneider-Kamp
QBF Encoding of Temporal Properties and QBF-Based Verification . . . . 224
Wenhui Zhang
Proof Theory
Introducing Quantified Cuts in Logic with Equality . . . . . . . . . . . . . . . . . . 240
Stefan Hetzl, Alexander Leitsch, Giselle Reis, Janos Tapolczai, and
Daniel Weller
Quati: An Automated Tool for Proving Permutation Lemmas . . . . . . . . . . 255
Vivek Nigam, Giselle Reis, and Leonardo Lima
A History-Based Theorem Prover for Intuitionistic Propositional Logic
Using Global Caching: IntHistGC System Description . . . . . . . . . . . . . . . . 262
Rajeev Goré, Jimmy Thomson, and Jesse Wu
MleanCoP: A Connection Prover for First-Order Modal Logic . . . . . . . . . . 269
Jens Otten
Modal Logic
Terminating Minimal Model Generation Procedures for Propositional
Modal Logics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 381
Fabio Papacchini and Renate A. Schmidt
Complexity
The Complexity of Theorem Proving in Circumscription and Minimal
Entailment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 403
Olaf Beyersdorff and Leroy Chew
XXVIII Table of Contents
Description Logics
Count and Forget: Uniform Interpolation of SHQ-Ontologies . . . . . . . . . . 434
Patrick Koopmann and Renate A. Schmidt
Knowledge Engineering for Large Ontologies with Sigma KEE 3.0 . . . . . . 519
Adam Pease and Stephan Schulz
1 Introduction
The classical definition of a computation in computer science uses to the model of a
Turing machine that recognizes a decidable language: once an input word is received,
the machine operates on it, and eventually terminates, accepting or rejecting the word.
Such a mode of operation corresponds to the use of computers for the solution of decid-
able problems, and there is no need to elaborate on the extensive research in theoretical
computer science about this model and issues like decidability and complexity. The
specification of a Turing machines is done by means of the language it recognizes.
Indeed, the specification of hardware and software systems that are input-output trans-
formers refers to the transformation they perform, for example “z = x·y” or “the vector
of strings is alphabetically sorted”.
The classical definition of a computation does not capture the mode of operation of
reactive systems [23]. Such systems maintain an on-going interaction with their envi-
ronment. Operating systems, ATMs, elevators, satellites – these are all reactive systems.
The computations of reactive systems need not terminate, and their specifications refer
to the on-going interaction of the system with its environment, for example “every re-
quest is eventually granted” or “two requests are never granted simultaneously”. Formal
methods for specification, verification, and design of reactive systems have been a very
active research area since the 80s.
S. Demri, D. Kapur, and C. Weidenbach (Eds.): IJCAR 2014, LNAI 8562, pp. 1–15, 2014.
c Springer International Publishing Switzerland 2014
2 G. Avni, O. Kupferman, and T. Tamir
1
Throughout this paper, we focus on pure strategies and pure deviations, as is the case for the
vast literature on cost-sharing games.
From Reachability to Temporal Specifications in Cost-Sharing Games 3
and a user may like to specify requirements like “all edges are operated by the same
provider” or “no edge operated by AT&T is followed by an edge operated by Verizon”.
Edges may also have different quality or security levels (e.g., “noisy channel”, “high-
bandwidth channel”, or “encrypted channel”), and again, users may like to specify their
preferences with respect to these properties. In planning or in production systems, nodes
of the network correspond to configurations, and edges correspond to the application of
actions. The objectives of the players are sequences of actions that fulfill a certain plan,
which is often more involved than just reachability [14]; for example “once the arm is
up, do not put it down until the block is placed”.
We extend network-formation games to a setting in which the players can specify
regular objectives. This involves two changes of the underlying setting: First, the edges
in the network are labeled by letters from a designated alphabet. Second, the objective
of each player is specified by a language over this alphabet. Each player should se-
lect a path labeled by a word in his objective language. Thus, if we view the network
as a nondeterministic weighted finite automaton (WFA) A, then the set of strategies
for a player with objective L is the set of accepting runs of A on some word in L.
Accordingly, we refer to our extension as automaton-formation games. As in classical
network-formation games, players share the cost of edges they use. Unlike the classical
game, the runs selected by the players need not be simple, thus a player may traverse
some edges several times. Edge costs are shared by the players, with the share being
proportional to the number of times the edge is traversed. This latter issue is the main
technical difference between automaton-formation and network-formation games, and
as we shall see, it is very significant.
Many variants of cost-sharing games and congestion games have been studied. A
generalization of the network-formation game of [2] in which players are weighted
and a player’s share in an edge cost is proportional to its weight is considered in [12],
where it is shown that the weighted game does not necessarily have a pure NE. In
a different type of congestion games, players’ payments depend on the resource they
choose to use, the set of players using this resource, or both [19,27,28,30]. In some of
these variants a NE is guaranteed to exist while in others it is not. All these variants are
different from automaton-formation games, where a player needs to select a multiset of
resources (namely, the edges he is going to traverse) rather than a single one.
We study the theoretical and practical aspects of automaton-formation games. In
addition to the general game, we consider classes of instances that have to do with the
network, the specifications, or their combination. Recall that the network can be viewed
as a WFA A. We consider the following classes of WFAs: (1) all-accepting, in which
all the states of A are accepting, thus its language is prefix closed (2) uniform costs, in
which all edges have the same cost, and (3) single letter, in which A is over a single-
letter alphabet. We consider the following classes of specifications: (1) single word,
where the language of each player is a single word, (2) symmetric, where all players
have the same objective. We also consider classes of instances that are intersections of
the above classes.
Each of the restricted classes we consider corresponds to a real-life variant of the
general setting. Let us elaborate below on single-letter instances. The language of an
automaton over a single letter {a} induces a subset of IN, namely the numbers k ∈ IN
4 G. Avni, O. Kupferman, and T. Tamir
The paper is based on our paper “Network-Formation Games with Regular Objec-
tives” [5]. Due to the lack of space, some proofs and examples are missing and can be
found in the full version.
2 Preliminaries
2.1 Automaton-Formation Games
A nondeterministic finite weighted automaton on finite words (WFA, for short) is a tuple
A = Σ, Q, Δ, q0 , F, c, where Σ is an alphabet, Q is a set of states, Δ ⊆ Q × Σ × Q
is a transition relation, q0 ∈ Q is an initial state, F ⊆ Q is a set of accepting states, and
c : Δ → IR is a function that maps each transition to the cost of its formation [29]. A
run of A on a word w = w1 , . . . , wn ∈ Σ ∗ is a sequence of states π = π 0 , π 1 , . . . , π n
such that π 0 = q0 and for every 0 ≤ i < n we have Δ(π i , wi+1 , π i+1 ). The run π is
accepting iff π n ∈ F . The length of π is n, whereas its size, denoted |π|, is the number
of different transitions in it. Note that |π| ≤ n.
An automaton-formation game (AF game, for short) between k selfish players is a
pair A, O, where A is a WFA over some alphabet Σ and O is a k-tuple of regular
languages over Σ. Thus, the objective of Player i is a regular language Li , and he needs
to choose a word wi ∈ Li and an accepting run of A on wi in a way that minimizes
his payments. The cost of each transition is shared by the players that use it in their
selected runs, where the share of a player in the cost of a transition e is proportional to
the number of times e is used by the player. Formally, The set of strategies for Player i
is Si = {π : π is an accepting run of A on some word in Li }. We assume that Si is not
empty. We refer to the set S = S1 × . . . × Sk as the set of profiles of the game.
Consider a profile P = π1 , π2 , . . . , πk . We refer to πi as a sequence of transitions.
Let πi = e1i , . . . , ei i , and let ηP : Δ → IN be a function that maps each transition in Δ
to the number of times it is traversed by all the strategies in P , taking into an account
several traversals in a single strategy. Denote by ηi (e) the number of times e is traversed
in πi , that is, ηi (e) = |{1 ≤ j ≤ i : eji = e}|. Then, ηP (e) = i=1...k ηi (e). The cost
of player i in the profile P is
ηi (e)
costi (P ) = c(e). (1)
e∈π
ηP (e)
i
For example, consider the WFA A depicted in Fig. 1. The label e1 : a, 1 on the
transition from q0 to q1 indicates that this transition, which we refer to as e1 , traverses
the letter a and its cost is 1. We consider a game between two players. Player 1’s ob-
jective is the language is L1 = {abi : i ≥ 2} and Player 2’s language is {ab, ba}.
Thus, S1 = {{e1 , e2 , e2 }, {e1, e2 , e2 , e2 }, . . .} and S2 = {{e3 , e4 }, {e1 , e2 }}. Con-
sider the profile P = {e1 , e2 , e2 }, {e3, e4 }, the strategies in P are disjoint, and
we have cost1 (P ) = 2 + 2 = 4, cost2 (P ) = 1 + 3 = 4. For the profile P =
{e1 , e2 , e2 }, {e1 , e2 }, it holds that η1 (e1 ) = η2 (e1 ) and η1 (e2 ) = 2 · η2 (e2 ). There-
fore, cost1 (P ) = 12 + 2 = 2 12 and cost2 (P ) = 12 + 1 = 1 21 .
We consider the following instances of AF games. Let G = A, O. We start with
instances obtained by imposing restrictions on the WFA A. In one-letter instances,
6 G. Avni, O. Kupferman, and T. Tamir
e2 : b, 3
e4 : b, 2 e3 : a, 2 e1 : a, 1
q3 q2 q0 q1
A is over a singleton alphabet, i.e., |Σ| = 1. When depicting such WFAs, we omit the
letters on the transitions. In all-accepting instances, all the states in A are accepting; i.e.,
F = Q. In uniform-costs instances, all the transitions in the WFA have the same cost,
which we normalize to 1. Formally, for every e ∈ Δ, we have c(e) = 1. We continue to
restrictions on the objectives in O. In single-word instances, each of the languages in O
consists of a single word. In symmetric instances, the languages in O coicide, thus the
players all have the same objective. We also consider combinations on the restrictions.
In particular, we say that A, O is weak if it is one-letter, all states are accepting, costs
are uniform, and objectives are single words. Weak instances are simple indeed – each
player only specifies a length of a path he should patrol, ending anywhere in the WFA,
where the cost of all transitions is the same. As we shall see, many of our hardness
results and lower bounds hold already for the class of weak instances.
he had with him all manner of vessels] Such informal tribute was
an acknowledgment of David’s suzerainty made in order to claim
David’s protection in war. Compare the action of Asa (1 Kings xv. 18,
19) and of Ahaz (2 Kings xvi. 7, 8). In all three cases the policy was
the same, i.e. to acknowledge a distant suzerain and so gain the
benefit of a valuable alliance while losing the minimum of
independence.
scribe] margin, secretary. See 2 Kings xii. 10, xviii. 18, xxii. 3;
compare 2 Kings xxv. 19, a passage which suggests that there was
a second scribe with military duties. The first, the king’s scribe, would
formulate the king’s orders and conduct his correspondence with
foreign powers. Shavsha’s sons held the office in the reign of
Solomon, 1 Kings iv. 3.
chief about the king] Literally, the chief at the kings hand, i.e.
formed the executive to carry out his commands; compare
Nehemiah xi. 24. In 2 Samuel viii. 18 (Revised Version) David’s sons
are described as priests, a statement which is in all probability
correct, but which the Chronicler, following the later theory of the
priesthood, could not accept (see Introduction, pp. xli. f.).
Chapter XIX.
1‒19 (= 2 Samuel x. 1‒19).
War with the Ammonites and their Aramean Allies.
Chronicles here omits the story of David’s kindness in seeking
out and befriending Mephibosheth (Merib-baal) the son of Jonathan
(2 Samuel ix.), because he has ignored the story of David’s relations
with Saul. Further the Court History of David which occupies an
important place in 2 Samuel is passed over altogether in Chronicles
Consequently the shameful episode of Bath-sheba, and the rebellion
of Absalom vanish from the account of David. It is obvious how
greatly the presentation of David’s life and character is affected by
these omissions. Yet from his point of view the Chronicler is right in
passing these matters by in silence. He was concerned to present
David essentially as the founder of the religious life of Israel as a
kingdom and of the Temple as an institution of religion.
² Hebrew Abshai.
10. he chose of all the choice men] The Syrians were the more
formidable because of the chariots they had; Joab therefore opposed
to them the flower of his army.
16‒19.
The End of the Aramean War.
Chapter XX.
1‒3 (= 2 Samuel xi. 1, xii. 26‒31).
The Subjugation of Ammon.
Between the two sections of this chapter the Chronicler omits the
account of the rebellions of Absalom and of Sheba, and the story of
the Gibeonite vengeance on the house of Saul (2 Samuel xiii. i‒xxi.
14).
Dan] The modern Tell el-Kādī, about forty minutes distance from
Bāniās (Paneas), north of Lake Huleh (Waters of Merom). For its
original name Laish, see Judges xviii. 28.