Get ebook downloads in full at ebookmeta.

com

Computer Science Theory and Applications 17th

International Computer Science Symposium in Russia
CSR 2022 Virtual Event June 29 July 1 2022 Lecture
Notes in Computer Science 13296 Alexander S.
Kulikov (Editor)
https://ebookmeta.com/product/computer-science-theory-and-
applications-17th-international-computer-science-symposium-
in-russia-csr-2022-virtual-event-
june-29-july-1-2022-lecture-notes-in-computer-
science-13296-alexander-s-kuliko/

OR CLICK BUTTON

DOWNLOAD NOW

Explore and download more ebook at https://ebookmeta.com

 Recommended digital products (PDF, EPUB, MOBI) that
you can download immediately if you are interested.

Computer Science Theory and Applications 16th

International Computer Science Symposium in Russia CSR
2021 Sochi Russia June 28 July 2 2021 1st Edition Rahul
Santhanam
https://ebookmeta.com/product/computer-science-theory-and-
applications-16th-international-computer-science-symposium-in-russia-
csr-2021-sochi-russia-june-28-july-2-2021-1st-edition-rahul-santhanam/
ebookmeta.com

Integer Programming and Combinatorial Optimization 23rd

International Conference IPCO 2022 Eindhoven The
Netherlands June 27 29 2022 Proceedings Lecture Notes in
Computer Science 13265 Karen Aardal (Editor)
https://ebookmeta.com/product/integer-programming-and-combinatorial-
optimization-23rd-international-conference-ipco-2022-eindhoven-the-
netherlands-june-27-29-2022-proceedings-lecture-notes-in-computer-
science-13265-karen-aardal/
ebookmeta.com

Computer Information Systems and Industrial Management:

21st International Conference, CISIM 2022, Barranquilla,
Colombia, July 15–17, 2022, Proceedings (Lecture Notes in
Computer Science, 13293) Khalid Saeed (Editor)
https://ebookmeta.com/product/computer-information-systems-and-
industrial-management-21st-international-conference-
cisim-2022-barranquilla-colombia-july-15-17-2022-proceedings-lecture-
notes-in-computer-science-132/
ebookmeta.com

Machine Learning Bookcamp Build a portfolio of real life

projects 1st Edition Alexey Grigorev

https://ebookmeta.com/product/machine-learning-bookcamp-build-a-
portfolio-of-real-life-projects-1st-edition-alexey-grigorev-2/

ebookmeta.com
Zane s Rebel A Curvy Girl Instalove Romance Silver Spoon
Falls 1st Edition Loni Nichole

https://ebookmeta.com/product/zane-s-rebel-a-curvy-girl-instalove-
romance-silver-spoon-falls-1st-edition-loni-nichole-2/

ebookmeta.com

The OTA s Guide to Documentation Writing SOAP Notes 5th

Edition Marie Morreale Otr

https://ebookmeta.com/product/the-ota-s-guide-to-documentation-
writing-soap-notes-5th-edition-marie-morreale-otr/

ebookmeta.com

Charm Destine Academy 7 1st Edition Sara Snow [Snow

https://ebookmeta.com/product/charm-destine-academy-7-1st-edition-
sara-snow-snow/

ebookmeta.com

A Night of Sin and Corruption: A Deadly Sins Anthology

(Vices and Hedonism) 1st Edition Indie Black

https://ebookmeta.com/product/a-night-of-sin-and-corruption-a-deadly-
sins-anthology-vices-and-hedonism-1st-edition-indie-black/

ebookmeta.com

Applied Combinatorics 2017 Edition Edition Keller Mitchel

T Trotter William T

https://ebookmeta.com/product/applied-combinatorics-2017-edition-
edition-keller-mitchel-t-trotter-william-t/

ebookmeta.com
Swarm Intelligence: Foundation, Principles, and
Engineering Applications 1st Edition Abhishek Sharma

https://ebookmeta.com/product/swarm-intelligence-foundation-
principles-and-engineering-applications-1st-edition-abhishek-sharma/

ebookmeta.com
 Alexander S. Kulikov
Sofya Raskhodnikova (Eds.)
LNCS 13296

Computer Science –
Theory and Applications
17th International Computer Science Symposium in Russia, CSR 2022
Virtual Event, June 29 – July 1, 2022
Proceedings
Lecture Notes in Computer Science 13296

Founding Editors
Gerhard Goos
Karlsruhe Institute of Technology, Karlsruhe, Germany
Juris Hartmanis
Cornell University, Ithaca, NY, USA

Editorial Board Members

Elisa Bertino
Purdue University, West Lafayette, IN, USA
Wen Gao
Peking University, Beijing, China
Bernhard Steffen
TU Dortmund University, Dortmund, Germany
Moti Yung
Columbia University, New York, NY, USA
More information about this series at https://link.springer.com/bookseries/558
Alexander S. Kulikov ·
Sofya Raskhodnikova (Eds.)

Computer Science –
Theory and Applications
17th International Computer Science Symposium in Russia, CSR 2022
Virtual Event, June 29 – July 1, 2022
Proceedings
Editors
Alexander S. Kulikov Sofya Raskhodnikova
Steklov Institute of Mathematics Boston University
St. Petersburg, Russia Boston, MA, USA

ISSN 0302-9743 ISSN 1611-3349 (electronic)

Lecture Notes in Computer Science
ISBN 978-3-031-09573-3 ISBN 978-3-031-09574-0 (eBook)
https://doi.org/10.1007/978-3-031-09574-0

© Springer Nature Switzerland AG 2022

This work is subject to copyright. All rights are reserved by the Publisher, whether the whole or part of the
material is concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation,
broadcasting, reproduction on microfilms or in any other physical way, and transmission or information
storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now
known or hereafter developed.
The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication
does not imply, even in the absence of a specific statement, that such names are exempt from the relevant
protective laws and regulations and therefore free for general use.
The publisher, the authors, and the editors are safe to assume that the advice and information in this book are
believed to be true and accurate at the date of publication. Neither the publisher nor the authors or the editors
give a warranty, expressed or implied, with respect to the material contained herein or for any errors or
omissions that may have been made. The publisher remains neutral with regard to jurisdictional claims in
published maps and institutional affiliations.

This Springer imprint is published by the registered company Springer Nature Switzerland AG
The registered company address is: Gewerbestrasse 11, 6330 Cham, Switzerland
 Preface

This volume contains the papers presented at CSR 2022, the 17th International Computer
Science Symposium in Russia, held online during June 29 – July 1, 2022. CSR covers
a wide range of areas in theoretical computer science and its applications. Initially,
CSR 2022 was planned as a satellite event for the International Congress of Mathemati-
cians (ICM) in St. Petersburg, Russia. However, as the Program Committee (PC) was
starting its deliberations after completing submission reviews, Russia attacked Ukraine.
As a result, ICM and CSR 2022 were moved online. Many PC members expressed dis-
may at the attack and three PC members resigned. Others chose to continue their work,
but many wanted to emphasize that they did not support or condone the actions of the
Russian government against Ukrainian people.
We received 51 submissions, and out of these the Program Committee selected 21
papers for presentation at the symposium and for publication in the proceedings. Each
submission was reviewed by at least three Program Committee members. Submissions
by Program Committee members were reviewed by at least four other members of the
Program Committee.
The opening lecture at CSR 2022 was given by Umesh Vazirani (University of
California at Berkeley), the closing lecture was given by Mark Braverman (Princeton
University). Three invited plenary lectures were given by Irit Dinur (Weizmann Institute
of Science), Jelani Nelson (University of California at Berkeley), and Mary Wootters
(Stanford University).
Many people and organizations contributed to the smooth running and the success
of CSR 2022. In particular, our thanks go to

– all authors who submitted their work to CSR;

– the members of the Program Committee who graciously devoted their time and energy
to the evaluation process;
– the expert reviewers who helped us evaluate the papers;
– the invited speakers; and
– the members of the local Organizing Committee who made the conference possible.

May 2022 Alexander S. Kulikov

Organizing Committee Chair
Sofya Raskhodnikova
Program Committee Chair
 Organization

Program Committee
Alexander Belov University of Latvia, Latvia
Sergio Cabello University of Ljubljana, Slovenia
Michael Elkin Ben-Gurion University of the Negev, Israel
(Resigned on March 1, 2022)
Edith Elkind University of Oxford, UK
Nathanaël Fijalkow CNRS, LaBRI, University of Bordeaux, France
Fedor Fomin University of Bergen, Norway
Moses Ganardi Max Planck Institute for Software Systems,
Germany (Resigned on March 3, 2022)
Alexander Golovnev Georgetown University, USA
Prahladh Harsha Tata Institute of Fundamental Research, India
Oded Lachish Birkbeck, University of London, UK
Daniel Lokshtanov University of California, Santa Barbara, USA
Dániel Marx Max Planck Institute for Informatics, Germany
(Resigned on March 1, 2022)
Pierre Ohlmann University of Warsaw, Poland
Alexander Okhotin Saint Petersburg State University, Russia
Madhusudan Parthasarathy University of Illinois at Urbana-Champaign, USA
Vladimir Podolskii Steklov Mathematical Institute, Russia
Svetlana Puzynina Saint Petersburg State University, Russia
Ramyaa Ramyaa New Mexico Institute of Mining and Technology
Sofya Raskhodnikova Boston University, USA
Barna Saha University of California, Berkeley, USA
Jeffrey Shallit University of Waterloo, Canada
Alexander Shen LIRMM, CNRS and Université de Montpellier,
France, on leave from IITP RAS, Russia

Additional Reviewers
Some chose to remain anonymous. Chakraborty, Dibyayan
Anand, Ashwani Chen, Lijie
Babu, Jasine Chew, Leroy
Barenboim, Leonid Chou, Chi-Ning
Bergsträßer, Pascal Das, Bireswar
Bienvenu, Laurent Dorfman, Dani
Cenzer, Douglas Efremenko, Klim
viii Organization

Eiben, Eduard Parreaux, Julie

Erlebach, Thomas Paul, Erik
Gawrychowski, Pawel Pentus, Mati
Giannopoulos, Panos Perifel, Sylvain
Goldberg, Paul Peters, Spencer
Gurvich, Vladimir Pruhs, Kirk
Haviv, Ishay Pérez-Lantero, Pablo
Hirahara, Shuichi Raskin, Michael
Itsykson, Dmitry Renault, Marc
Kalimullin, Iskander Romashchenko, Andrei
Kavitha, Telikepalli
Rosenfeld, Matthieu
Knop, Alexander
Schlöter, Miriam
Kozachinskiy, Alexander
Seelbach Benkner, Louisa
Kumar, Mrinal
Selivanov, Victor
Köppl, Dominik
Li, Jason Sering, Leon
M. Sridharan, Ramanujan Serre, Olivier
Makarov, Vladislav Slivovsky, Friedrich
Meeks, Kitty Solomon, Shay
Michielini, Vincent Spoerhase, Joachim
Milovanov, Alexey Srinivasan, Srikanth
Mozes, Shay Suksompong, Warut
Nanashima, Mikito Trehan, Chhaya
Neiman, Ofer Vereshchagin, Nikolay
Neuen, Daniel Vinyals, Marc
Ochem, Pascal Wlodarczyk, Michal
Invited Talks
 Optimization-Friendly Generic Mechanisms Without
Money

Mark Braverman

Princeton University, Princeton, NJ 08540, USA

https://mbraverm.princeton.edu

Our goal is to develop a generic framework for converting modern gradient-descent

based optimization algorithms into mechanisms where inputs come from self-interested
agents.
We focus on aggregating preferences from n players in a context without money.
Special cases of this setting include voting, allocation of items by lottery, and matching.
Our key technical contribution is a new meta-algorithm we call APEX (Adaptive Pricing
Equalizing Externalities). The framework is sufficiently general to be combined with
any optimization algorithm that is based on local search. In the talk we outline the
algorithm, and open problem/research directions that it raises, with a particular focus
towards mechanism design + machine learning.
We discuss a special case of applying the framework to the problem of one-sided
allocation with lotteries. In this case, we obtain a strengthening of the 1979 result by
Hylland and Zeckhauser on allocation via a competitive equilibrium from equal incomes
(CEEI). The [HZ79] result posits that there is a (fractional) allocation and a set of item
prices such that the allocation is a competitive equilibrium given prices. We further
show that there is always a reweighing of the players’ utility values such that running the
standard unit-demand VCG with reweighed utilities leads to a HZ-equilibrium prices.
Interestingly, not all HZ competitive equilibria come from VCG prices.

Reference
[HZ79] Hylland, A., Zeckhauser, R.: The efficient allocation of individuals to positions.
J. Polit. Econ. 87(2), 293–314 (1979)
 Expanders in Higher Dimensions

Irit Dinur

Weizmann Institute of Science

Expander graphs have been studied in many areas of mathematics and in computer
science with versatile applications, including coding theory, networking, computational
complexity and geometry.
High-dimensional expanders are a generalization that has been studied in recent
years and their promise is beginning to bear fruit. In the talk, I will survey some pow-
erful local to global properties of high-dimensional expanders, and describe several
interesting applications, ranging from convergence of random walks to construction of
locally testable codes that prove the c3 conjecture (namely, codes with constant rate,
constant distance, and constant locality).
 Private Frequency Estimation via Projective Geometry

Jelani Nelson

UC Berkeley, Berkeley, CA 94705, USA

https://people.eecs.berkeley.edu/~minilek

We propose a new algorithm ProjectiveGeometryResponse (PGR) for locally dif-

ferentially private (LDP) frequency estimation. For a universe size of k and with n users,
our ε-LDP algorithm has communication cost log2 k bits in the private coin setting and
εlog2 e + O(1) in the public coin setting, and has computation cost O(n + k exp(ε)logk)
for the server to approximately reconstruct the frequency histogram, while achieving
optimal privacy/utility tradeoff, including optimality of the leading constant factor.
Our empirical evaluation shows a speedup of over 50x over PI-RAPPOR [FT21],
while using approximately 75x less memory for practically relevant parameter set-
tings. In addition, the running time of our algorithm is within an order of magnitude of
HadamardResponse [ASZ19] and RecursiveHadamardResponse [CKO20] which
have significantly worse reconstruction error. Our new algorithm is based on using Pro-
jective Planes over a finite field to define a small collection of sets that are close to
being pairwise independent and a dynamic programming algorithm for approximate
histogram reconstruction on the server side. We also give an extension of PGR, which
we call HybridProjectiveGeometryResponse, that allows trading off computation
time with utility smoothly.
Joint work with Vitaly Feldman (Apple), Huy Le Nguyen (Northeastern), and Kunal
Talwar (Apple). This work is to appear in ICML 2022.

References
[ASZ19] Acharya, J., Sun, Z., Zhang, H.: Hadamard response: estimating distributions
privately, efficiently, and with little communication. In: Proceedings of the 22nd Inter-
national Conference on Artificial Intelligence and Statistics (AISTATS), pp. 1120–129
(2019)
[CKO20] Chen, W.-N., Kairouz, p., Özgür, A.: Breaking the communication-privacy-
accuracy trilemma. In: Proceedings of the 32nd Annual Conference on Advances in
Neural Information Processing Systems (NeurIPS) (2020)
[FT21] Feldman, V., Talwar, K.: Lossless compression of efficient private local ran-
domizers. In: Proceedings of the 38th International Conference on Machine Learning
(ICML), pp. 3208–3219 (2021)
 Contents

Parameterized Algorithms for Finding Highly Connected Solution . . . . . . . . . . . . 1

Ankit Abhinav, Susobhan Bandopadhyay, Aritra Banik, and Saket Saurabh

Coloring a Dominating Set Without Conflicts: q-Subset Square Coloring . . . . . . 17

V. P. Abidha, Pradeesha Ashok, Avi Tomar, and Dolly Yadav

Quotient Structures and Groups Computable in Polynomial Time . . . . . . . . . . . . . 35

Pavel Alaev

Parameterized Complexity of List Coloring and Max Coloring . . . . . . . . . . . . . . . 46

Bardiya Aryanfard and Fahad Panolan

Eternal Vertex Cover on Bipartite Graphs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64

Jasine Babu, Neeldhara Misra, and Saraswati Girish Nanoti

Non-crossing Shortest Paths in Undirected Unweighted Planar Graphs

in Linear Time . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 77
Lorenzo Balzotti and Paolo G. Franciosa

Lossy Kernelization of Same-Size Clustering . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 96

Sayan Bandyapadhyay, Fedor V. Fomin, Petr A. Golovach,
Nidhi Purohit, and Kirill Siminov

Output Sensitive Fault Tolerant Maximum Matching . . . . . . . . . . . . . . . . . . . . . . . . 115

Niranka Banerjee, Manoj Gupta, Venkatesh Raman, and Saket Saurabh

Bounds for Synchronizing Markov Decision Processes . . . . . . . . . . . . . . . . . . . . . . 133

Laurent Doyen and Marie van den Bogaard

Parameterized Complexity of Set-Restricted Disjoint Paths on Chordal

Graphs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 152
Petr A. Golovach, Fahad Panolan, Ashutosh Rai, and Saket Saurabh

Discrete Versions of the KKM Lemma and Their PPAD-Completeness . . . . . . . . 170

Alexander Grishutin and Daniil Musatov

The Fast Algorithm for Online k-server Problem on Trees . . . . . . . . . . . . . . . . . . . 190

Kamil Khadiev and Maxim Yagafarov
xvi Contents

Finite Ambiguity and Finite Sequentiality in Weighted Automata

over Fields . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 209
Peter Kostolányi

New Bounds for the Flock-of-Birds Problem . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 224

Alexander Kozachinskiy

Heterogeneous Multi-commodity Network Flows over Time . . . . . . . . . . . . . . . . . 238

Yifen Li, Xiaohui Bei, Youming Qiao, Dacheng Tao, and Zhiya Chen

On the Determinization of Event-Clock Input-Driven Pushdown Automata . . . . 256

Mizuhito Ogawa and Alexander Okhotin

The GKK Algorithm is the Fastest over Simple Mean-Payoff Games . . . . . . . . . . 269
Pierre Ohlmann

Partial Vertex Cover on Graphs of Bounded Degeneracy . . . . . . . . . . . . . . . . . . . . 289

Fahad Panolan and Hannane Yaghoubizade

Abelian Repetition Threshold Revisited . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 302

Elena A. Petrova and Arseny M. Shur

Characterizing Level One in Group-Based Concatenation Hierarchies . . . . . . . . . 320

Thomas Place and Marc Zeitoun

How Much Randomness is Needed to Convert MA Protocols to AM

Protocols? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 338
Nikolay Vereshchagin

Author Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 351

 Parameterized Algorithms for Finding
Highly Connected Solution

Ankit Abhinav1 , Susobhan Bandopadhyay1(B) , Aritra Banik1 ,

and Saket Saurabh2
1
National Institute of Science, Education and Research,
An OCC of Homi Bhabha National Institute, Bhubaneswar 752050, Odisha, India
{ankit.abhinav,susobhan.bandopadhyay,aritra}@niser.ac.in
2
The Institute of Mathematical Sciences, HBNI, Chennai, India
[email protected]

Abstract. To introduce our question and the parameterization, consider

the classical Vertex Cover problem. In this problem, the input is a graph
G on n vertices and a positive integer , and the goal is to find a vertex sub-
set S of size at most  such that G − S is an independent set. Further, we
want that G[S] is highly connected. That is, G[S] should be n − k edge-
connected. Clearly, the problem is NP-complete, as substituting k = n−1,
we obtain the Connected Vertex Cover problem. A simple observa-
tion also shows that the problem admits an algorithm with running time
nO(k) . Since the problem is polynomial-time solvable for every fixed inte-
ger k, a natural parameter is the integer k. In all the problems we consider,
the parameter is k, and the goal is to find a solution S of size at most ,
such that G[S] is n − k edge-connected and G − S satisfies a property. We
show that this version of well-known problems such as Vertex Cover,
Feedback Vertex Set, Odd Cycle Transversal and Multiway Cut
admit an algorithm with running time f (k) · nO(1) , that is, they are FPT
with the parameter k. One of our main subroutines to obtain these algo-
rithms is an FPT algorithm for n − k edge connected Steiner Subgraph,
which could be of an independent interest. Finally, we also show that such
an algorithm is not possible for Multicut.

Keywords: subset problems · parameterized algorithms · connectivity

1 Introduction

Vertex deletion (subset) problems form an important sub-area of graph optimiza-

tion problems. An input to a prototype vertex deletion problem consists of a graph
G and an integer  and the objective is to find a vertex subset S of size at most  such
that G − S satisfies a property, such as being an edgeless graph (Vertex Cover),
an acyclic graph (FVS), a bipartite graph(OCT), a chordal graph(CVD), a planar
graph (PVD), and a (topological) minor-free graph. In literature, several variants
of these classical vertex deletion problems are considered. The most notable ones
c Springer Nature Switzerland AG 2022
A. S. Kulikov and S. Raskhodnikova (Eds.): CSR 2022, LNCS 13296, pp. 1–16, 2022.
https://doi.org/10.1007/978-3-031-09574-0_1
Other documents randomly have
different content
on the 23d, his cavalry being near Usagre. This abandonment of the
royal road to Seville was a well-considered movement. The country
through which Soult passed being more fruitful and open, he could
draw greater advantage from his superior cavalry; the mountains
behind him were so strong he had nothing to fear from an attack;
and by Belalcazar and Almaden, he could maintain a communication
with La Mancha, from whence he expected Drouet’s division. The
road of Guadalcanal was in his rear, by which he could draw
reinforcements from Cordoba and from the fourth corps, and
meanwhile the allies durst not venture to expose their left flank by
marching on Monasterio.
From Llerena, a detachment was sent to drive away a Spanish
Partizan corps which had cut his communications with Guadalcanal,
and at the same time Latour Maubourg was directed to scour the
country beyond Usagre; this led to an action; for that town, built upon
a hill, and covered towards Los Santos by a river with steep and
rugged banks, had only the one outlet by the bridge on that side, and
when Latour Maubourg approached, Lumley retired across the river.
The French light cavalry then marched along the right bank, with the
intention of crossing lower down and thus covering the passage of
the heavy horsemen; but before they could effect this object, general
Bron rashly passed the river with two regiments of dragoons, and
drew up in line just beyond the bridge. Lumley was lying close
behind a rising ground, and when the French regiments had
advanced a sufficient distance, Lefebre’s guns opened on them, and
the third, and fourth dragoon guards, charged them in front while
Madden’s Portuguese fell on their flank. They were overthrown at the
first shock, and fled towards the bridge, but that being choked by the
remainder of the cavalry advancing to their support, the fugitives,
turned to the right and left, endeavouring to save themselves
amongst some gardens situated on the banks of the river; there they
were pursued and sabred until the French on the opposite side,
seeing their distress, opened a fire of carbines and artillery that
obliged the British to discontinue the attack. Forty killed, above a
hundred wounded, and eighty prisoners were the fruits of this brilliant
action of general Lumley’s, which terminated Beresford’s operations,
for the miserable state to which the Regency had reduced the
Portuguese army imperatively called for the marshal’s presence.
General Hill, who had returned to Portugal, then re-assumed the
command of the second division, amidst the eager rejoicings of the
troops, and lord Wellington directed the renewed siege of Badajos in
person.

O B S E RVAT I O N S .

No general ever gained a great battle with so little increase of

military reputation as marshal Beresford. His personal intrepidity and
strength, qualities so attractive for the multitude, were conspicuously
displayed, yet the breath of his own army withered his laurels, and
his triumph was disputed by the very soldiers who followed his car.
Their censures have been reiterated, without change and without
abatement, even to this hour; and a close examination of his
operations, while it detects many ill-founded objections, and others
tainted with malice, leaves little doubt that the general feeling was
right.
When he had passed the Guadiana and driven the fifth corps upon
Guadalcanal, the delay that intervened, before he invested Badajos,
was unjustly attributed to him: it was lord Wellington’s order, resulting
from the tardiness of the Spanish generals, that paralyzed his
operations. But when the time for action arrived, the want of concert
in the investment, and the ill-matured attack on San Christoval
belonged to Beresford’s arrangements; and he is especially
responsible in reputation for the latter, because captain Squires
personally represented the inevitable result, and his words were
unheeded.
During the progress of the siege, either the want of correct
intelligence, or a blunted judgement, misled the marshal. It was
remarked that, at all times, he too readily believed the idle tales of
distress and difficulties in the French armies, with which the spies
generally, and the deserters always, interlarded their information.
Thus he was incredulous of Soult’s enterprise, and that marshal was
actually over the Morena before the orders were given for the
commencing of the main attack of the castle of Badajos. However,
the firmness with which Beresford resisted the importunities of the
engineers to continue the siege, and the quick and orderly removal
of the stores and battering-train, were alike remarkable and
praiseworthy. It would have been happy if he had shewn as much
magnanimity in what followed.
When he met Blake and Castaños at Valverde, the alternative of
fighting or retiring behind the Guadiana was the subject of
consideration. The Spanish generals were both in favour of giving
battle. Blake, who could not retire the way he had arrived, without
danger of having his march intercepted, was particularly earnest to
fight; affirming that his troops, who were already in a miserable state,
would disperse entirely if they were obliged to enter Portugal.
Castaños was of the same opinion. Beresford also argued that it was
unwise to relinquish the hope of taking Badajos, and ungenerous to
desert the people of Estremadura; that a retreat would endanger
Elvas, lay open the Alemtejo, and encourage the enemy to push his
incursions further, which he could safely do, having such a fortress
as Badajos with its bridge over the Guadiana, in his rear; a battle
must then be fought in the Alemtejo with fewer troops and after a
dispiriting retreat; there was also a greater scarcity of food in the
Portuguese than in the Spanish province, and, finally, as the weather
was menacing, the Guadiana might again rise before the stores were
carried over, when the latter must be abandoned, or the army
endangered to protect their passage.
But these plausible reasons were but a mask; the true cause why
the English general adopted Blake’s proposals was the impatient
temper of the British troops. None of them had been engaged in the
battles under lord Wellington. At Busaco the regiments of the fourth
division were idle spectators on the left, as those of the second
division were on the right, while the action was in the centre. During
Massena’s retreat they had not been employed under fire, and the
combats of Sabugal and Fuentes Onoro had been fought without
them. Thus a burning thirst for battle was general, and Beresford had
not the art either of conciliating or of exacting the confidence of his
troops. It is certain that if he had retreated, a very violent and unjust
clamour would have been raised against him, and this was so
strongly and unceremoniously represented to him, by an officer on
his own staff, that he gave way. These are what may be termed the
moral obstacles of war. Such men as lord Wellington or sir John
Moore can stride over them, but to second-rate minds they are
insuperable. Practice and study may make a good general as far as
the handling of troops and the designing of a campaign, but that
ascendancy of spirit which leads the wise, and controls the insolence
of folly, is a rare gift of nature.
Beresford yielded with an unhappy flexibility to the clamour of the
army and the representations of Blake, for it is unquestionable that
the resolution to fight was unwarrantable on any sound military
principle. We may pass over the argument founded upon the taking
of Badajos, because neither the measures nor the means of the
English general promised the slightest chance of success; the siege
would have died away of itself in default of resources to carry it on.
The true question to consider was, not whether Estremadura should
be deserted or Badajos abandoned, but whether lord Wellington’s
combinations and his great and well considered design for the
deliverance of the Peninsula, should be ruined and defaced at a
blow. To say that the Alemtejo could not have been defended until
the commander-in-chief arrived from the north with reinforcements
was mere trifling. Soult, with twenty or even thirty thousand men,
durst not have attempted the siege of Elvas in the face of twenty-four
thousand men such as Beresford commanded. The result of the
battle of Fuentes Onoro was known in the English and in the French
camps, before Beresford broke up from Badajos, hence he was
certain that additional troops would soon be brought down to the
Guadiana; indeed, the third and seventh divisions were actually at
Campo Mayor the 23d of May. The danger to the Alemtejo was,
therefore, slight, and the necessity of a battle being by no means
apparent, it remains to analyze the chances of success.
Soult’s numbers were not accurately known, but it was ascertained
that he had not less than twenty thousand veteran troops. He had
also a great superiority of cavalry and artillery, and the country was
peculiarly suitable for these arms; the martial character of the man
was also understood. Now the allies could bring into the field more of
infantry by ten thousand than the French, but they were of various
tongues, and the Spanish part ill armed, starving, and worn out with
fatigue, had been repeatedly and recently defeated by the very
troops they were going to engage. The French were compact, swift
of movement, inured to war, used to act together, and under the
command of one able and experienced general. The allied army was
unwieldy, each nation mistrusting the other, and the whole without
unity of spirit, or of discipline, or of command. On what, then, could
marshal Beresford found his hopes of success? The British troops.
The latter were therefore to be freely used. But was it a time to risk
the total destruction of two superb divisions and to encounter a
certain and heavy loss of men, whose value he knew so well when
he calculated upon them alone for victory in such circumstances?
To resolve on battle was, however, easier than to prepare for it
with skill. Albuera, we have seen, was the point of concentration.
Colonel Colborne’s brigade did not arrive until the 14th, and these
was no certainty that it could arrive before the enemy did. Blake did
not arrive until three in the morning of the 16th. The fourth division
not until six o’clock. Kemmis with three fine British regiments, and
Maddens cavalry, did not come at all. These facts prove that the
whole plan was faulty, it was mere accident that a sufficient force to
give battle was concentrated. Beresford was too late, and the
keeping up the investment of Badajos, although laudable in one
sense, was a great error; it was only an accessary, and yet the
success of the principal object was made subservient to it. If Soult,
instead of passing by Villa Franca, in his advance, had pushed
straight on from Los Santos to Albuera, he would have arrived the
15th, when Beresford had not much more than half his force in
position; the point of concentration would then have been lost, and
the allies scattered in all directions. If the French had even continued
their march by Solano instead of turning upon Albuera, they must
inevitably have communicated with Badajos, unless Beresford had
fought without waiting for Blake, and without Kemmis’s brigade. Why,
then, did the French marshal turn out of the way to seek a battle, in
preference to attaining his object without one? and why did he
neglect to operate by his right or left until the unwieldy allied army
should separate or get into confusion, as it inevitably would have
done? Because the English general’s dispositions were so faulty that
no worse error could well be expected from him, and Soult had every
reason to hope for a great and decided victory; a victory which would
have more than counterbalanced Massena’s failure. He knew that
only one half of the allied force was at Albuera on the 15th, and
when he examined the ground, every thing promised the most
complete success.
Marshal Beresford had fixed upon and studied his own field of
battle above a month before the action took place, and yet occupied
it in such a manner as to render defeat almost certain; his infantry
were not held in hand, and his inferiority in guns and cavalry was not
compensated for by entrenchments. But were any other proofs of
error wanting, this fact would suffice, he had a greater strength of
infantry on a field of battle scarcely four miles long, and three times
the day was lost and won, the allies being always fewest in number
at the decisive point. It is true that Blake’s conduct was very
perplexing; it is true that general William Stewart’s error cost one
brigade, and thus annihilated the command of colonel Colborne, a
man capable of turning the fate of a battle even with fewer troops
than those swept away from him by the French cavalry: but the
neglect of the hill beyond the Albuera, fronting the right of the
position, was Beresford’s own error and a most serious one; so also
were the successive attacks of the brigades, and the hesitation
about the fourth division. And where are we to look for that
promptness in critical moments which marks the great commander?
It was colonel Hardinge that gave the fourth division and
Abercrombie’s brigade orders to advance, and it was their
astounding valour in attack, and the astonishing firmness of
Houghton’s brigade in defence that saved the day; the person of the
general-in-chief was indeed seen every where, a gallant soldier! but
the mind of the great commander was seen no where.
Beresford remained master of the field of battle, but he could not
take Badajos, that prize was the result of many great efforts, and
many deep combinations by a far greater man: neither did he clear
Estremadura, for Soult maintained positions from Llerena to Usagre.
What then did he gain? The power of simulating a renewal of the
siege, and holding his own cantonments on the left bank of the
Guadiana; I say simulating, for, if the third and seventh divisions had
not arrived from Beira, even the investment could not have been
completed. These illusive advantages he purchased at the price of
seven thousand men. Now lord Wellington fought two general and
several minor actions, with a smaller loss, and moreover turned
Massena and seventy thousand men out of Portugal!
Such being the fruit of victory, what would have been the result of
defeat? There was no retreat, save by the temporary bridge of
Jerumenha, but, had the hill on the right been carried in the battle,
the Valverde road would have been in Soult’s possession, and the
line of retreat cut; and, had it been otherwise, Beresford, with four
thousand victorious French cavalry at his heels, could never have
passed the river. Back, then, must have come the army from the
north, the Lines of Lisbon would have been once more occupied—a
French force fixed on the south of the Tagus—Spain ruined—
Portugal laid prostrate—England in dismay. Could even the genius of
lord Wellington have recovered such a state of affairs? And yet, with
these results, the terrible balance hung for two hours, and twice
trembling to the sinister side, only yielded at last to the superlative
vigour of the fuzileers. The battle should never have been fought.
The siege of Badajos could not have been renewed without
reinforcements, and, with them, it could have been renewed without
an action, or at least without risking an unequal one.
But would even the bravery of British soldiers have saved the day,
at Albuera, if the French general had not also committed great
errors. His plan of attack and his execution of it, up to the moment
when the Spanish line fell back in disorder, cannot be too much
admired; after that, the great error of fighting in dense columns being
persisted in beyond reason, lost the fairest field ever offered to the
arms of France. Had the fifth corps opened out while there was time
to do so, that is, between the falling back of the Spaniards and the
advance of Houghton’s brigade, what on earth could have saved
Beresford from a total defeat? The fire of the enemy’s columns alone
destroyed two-thirds of his British troops; the fire of their lines would
have swept away all!
 It has been said that Latour Maubourg and Godinot did not second
Soult with sufficient vigour; the latter certainly did not display any
great energy, but the village was maintained by Alten’s Germans,
who were good and hardy troops, and well backed up by a great
body of Portuguese. Latour Maubourg’s movements seem to have
been objected to without reason. He took six guns, sabred many
Spaniards, and overthrew a whole brigade of the British, without
ceasing to keep in check their cavalry. He was, undoubtedly, greatly
superior in numbers, but general Lumley handled the allied
squadrons with skill and courage, and drew all the advantage
possible from his situation, and, in the choice of that situation, none
can deny ability to marshal Beresford. The rising ground behind the
horsemen, the bed of the Aroya in their front, the aid of the horse-
artillery, and the support of the fourth division, were all
circumstances of strength so well combined that nothing could be
better, and they dictated Latour Maubourg’s proceedings, which
seem consonant to true principles. If he had charged in mass, under
the fire of Lefebre’s guns, he must have been thrown into confusion
in passing the Aroya at the moment when the fourth division,
advancing along the slopes, would have opened a musketry on his
right flank; Lumley could then have charged, or retired up the hill,
according to circumstances. In this case, great loss might have been
sustained, and nothing very decisive could have accrued to the
advantage of the French, because no number of cavalry, if
unsustained by infantry and artillery, can make a serious impression
against the three arms united.
On the other hand, a repulse might have been fatal not only to
himself but to the French infantry on the hill, as their left would have
been open to the enterprises of the allied cavalry. If Latour Maubourg
had stretched away to his own left, he would, in like manner, have
exposed the flank of Soult’s infantry, and his movements would have
been eccentric, and contrary to sound principles; and, (in the event
of a disaster to the corps on the hill, as really happened,) destructive
to the safety of the retreating army. By keeping in mass on the plain,
and detaching squadrons from time to time, as favourable
opportunities offered for partial charges, he gained, as we have
seen, great advantages during the action, and kept his troopers well
in hand for the decisive moment; finally, he covered the retreat of the
beaten infantry. Still it may be admitted that, with such superior
numbers, he might have more closely pressed Lumley.
When Soult had regained the hills at the other side of the Albuera,
the battle ceased, each side being, as we have seen, so hardly
handled that neither offered to renew the fight. Here was the greatest
failure of the French commander; he had lost eight thousand men,
but he had still fifteen thousand under arms, and his artillery and his
cavalry were comparatively untouched. On the side of the allies, only
fifteen hundred British infantry were standing; the troops were
suffering greatly from famine; the Spaniards had been feeding on
horseflesh, and were so extenuated by continual fatigue and misery,
that, for several days previous to the battle, they had gone over in
considerable numbers even to the French, hoping thus to get food:
these circumstances should be borne in mind, when reflecting on
their conduct in the battle; under such a commander as Blake, and,
while enduring such heavy privations, it was a great effort of
resolution, and honourable to them that they fought at all. Their
resistance feeble, when compared to the desperate valour of the
British, was by no means weak in itself or infirm; nor is it to be
wondered at that men so exhausted and so ill-managed should have
been deaf to the call of Beresford, a strange general, whose
exhortations they probably did not understand. When the fortune of
the day changed they followed the fuzileers with alacrity, and at no
period did they give way with dishonour.
Nevertheless, all circumstances considered, they were not and
could not be equal to a second desperate struggle, a renewed attack
on the 17th, would have certainly ended in favour of the French; and
so conscious was Beresford of this, that, on the evening of the 16th,
he wrote to lord Wellington, avowing that he anticipated a certain
and ruinous defeat the next day. The resolution with which he
maintained the position notwithstanding, was the strongest indication
of military talent he gave during the whole of his operations; had
Soult only persisted in holding his position with equal pertinacity,
Beresford must have retired. It was a great and decided mistake of
the French marshal not to have done so. There is nothing more
essential in war than a confident front; a general should never
acknowledge himself vanquished, for the front line of an army always
looks formidable, and the adversary can seldom see the real state of
what is behind. The importance of this maxim is finely indicated in
Livy, where he relates that, after a drawn battle, a god called out in
the night, the Etruscans had lost one man more than the Romans!
Hereupon the former retired, and the latter, remaining on the field,
gathered all the fruits of a real victory.
 PA P E R S R E L AT I N G T O T H E
FORMER VOLUMES.

Letter from major-general F. Ponsonby to colonel Napier.

After the very handsome manner in which you have mentioned my name, in
your account of the battle of Talavera, it may appear extraordinary that I should
trouble you with this letter; but my silence might be interpreted into the wish of
taking praise to myself which I do not deserve.
The whole of your account of the charge made by general Anson’s brigade is
substantially correct; you have given the reason for it, and the result; but there are
two points, in the detail, which are inaccurate. The first affecting the German
hussars; the other respecting myself.
The Germans, on the left of the twenty-third, could not reach the French
columns, from the impracticability of the ravine where they charged; this I
ascertained, by personal observation, the following day; the obstacle was much
less serious where the twenty-third attacked, headed by general Anson and
colonel Seymour. The mountain torrent, which gradually decreased as it
descended into the plain, was about thirty yards in front of the enemy, and the
twenty-third, though much broken in passing this obstacle, charged up to the
columns, and was repulsed, no rally could be attempted; but the right squadron,
under captain Drake, having an easier passage of the ravine, and no French
column immediately in front, passed through the intervals, and caused much
confusion, which, together with the delay occasioned by the charge, prevented the
masses of infantry which were in readiness on the French right flank, from joining
in the general attack on our line.
You will perceive that this account, which I believe to be the exact truth, does
not, in the slightest degree, affect the accuracy of your description of the
movement; but, if I am correct, it proves that the Germans were obliged to halt by
an insuperable difficulty, and that I had no particular merit in the execution of the
charge of the twenty-third.
Believe me
Very sincerely yours,
 F. Ponsonby.
Malta, Dec. 30, 1829.

Note sur la Situation actuelle de l’Espagne.

Rochefort, le Août, 1808.
1º. Les événemens inattendus du général Dupont sont une preuve de plus que
le succès de la guerre dépend de la prudence, de la bonne conduite, et de
l’expérience du général.
2º. A la seule lecture du rapport du colonel d’Affry, on avoit diviné tous les
événemens; après une perte aussi considérable, on ne peut être surpris que le roi
et les généraux jugent convenable de concentrer l’armée et d’évacuer Madrid.
En examinant avec attention, non les rapports mensongers des individus qui
parlent dans leur sens, mais les faits tels qu’ils se sont passés, on est convaincu:
premièrement, que le général Castaños n’avoit pas plus de vingt-cinq mille
hommes de troupe de ligne et de quinze mille paysans; un jour on sera à même
de vérifier ce qui sera avancé ici. Secondement, que si le général Dupont les eut
attaqués ou se fût battû; avec tout son corps réuni, il les eut complettement
défaits.
3º. On pense qu’on aura tout le tems d’évacuer les blessés de Madrid qui
arrivent à Aranda; il faudra occuper aussi longtems qu’il sera possible les hauteurs
de Buitrago, afin de donner le temps au maréchal Bessières de revenir de son
mouvement de Gallice; qu’il faut réorganiser la province de Burgos, les trois
Biscayes, et la province de Navarre; elle comprendront facilement que, dans ce
moment plus que jamais, elles doivent rester fidèles et se bien conduire sous
peine d’être traitées avec toute la rigueur de la guerre.
4º. On pense que l’armée doit être divisée en trois corps, le corps principal, ou
de centre, où commande le roi, qu’on porteroit à 30,000 hommes campé à Aranda;
le corps de droite, du maréchal Bessières d’environ 15 mille hommes faisant face
à ce qui pourroit arriver de Gallice ou d’Estramadura, occupant Valladolid par une
division, ayant une autre division intermédiaire avec le corps du centre, et une
troisième division de plus sur sa droite, selon les circonstances; enfin le corps de
gauche, ou d’Arragon destiné à maintenir la Navarre et le pays environnant,
occupant Logrono et Tudela et liant sa droite au corps du centre, par une division
qui au besoin renforceroit ce corps et devra maintenir Soria par un corps volant.
Le corps du centre, et le corps de droite doivent s’appuyer sur Burgos et le
corps d’Arragon doit avoir son appui sur Pampelune.
5º. Pour organiser le corps du centre dans ce bût, on croit qu’on doit le renforcer
de la brigade du 14me et 44me de ligne, 200 chevaux et 8 pièces de canon, qu’on
tireroit du corps devant Saragosse; de la brigade du général Mouton composée du
4me legère, 15me legère, du bataillon de Paris, et de huit pièces de canon; de la
brigade commandée par le maréchal Ney, et qui est déjà à une marche en avant
de Bayonne, composée du 43me, et du 51me de ligne, du 26me de chasseurs, et de
6 pièces de canon; enfin de 4 escadrons de marche de dragons et d’une régiment
Polonais de la garde; on réuniroit le 3me bataillon aux deux premiers de tous les
régimens d’infanterie, et on méleroit les jeunes soldats aux anciens.
On évalue à environ dix mille hommes de renfort que recevroit le corps du
centre, qui seroit alors composé: savoir des
18,000 qui le forment à présent 18,000
Du renfort évalué à 10,000
Le détachement du depôt du 4me legère, 15me legère.
14me, 44me, 43me, et 51me de ligne, le 2me et 12me legère rejoindront
insensiblement et porteront ce corps à 30,000 hommes.
Ces trente mille hommes ne sauroient être en meilleure mains, que sous les
ordres du maréchal Ney, hormis une réserve de 4 à 5 mille hommes destinés à la
garde du roi, et que le roi conserveroit auprès de sa personne et feroit marcher
avec le général Saligny, ou avec le général Savary quand il le jugeroit nécessaire.
Le corps du centre ce tiendrait à la hauteur d’Aranda, ses communications bien
assurées avec le maréchal Bessières à Valladolid, des têtes de pont bien établies
à Aranda et à Valladolid. Ce corps se nourrira par Burgos et devra non seulement
maintenir la tranquillité dans cette province, mais encore assurer ses
communications avec le corps de Saragosse qui occupera Tudela et Logrono.
Le corps du maréchal Bessières, fort de quinze mille hommes, devra occuper
Valladolid en faisant face à ce qui arrivera d’Estramadure et de Castille, ayant ses
trois divisions en échélons et se nourrissant de la province de Valladolid,
Placentia, et Leon.
On enverra le maréchal Moncey pour commander le corps du général Verdier, et
on chargera le maréchal du commandement de la Biscaye et de tous les derrières.
On estime qu’on peut retirer du camp sous Saragosse le 14me et 44me de ligne,
200 chevaux, et 8 pièces de canon, le reste doit être formé en trois divisions, et
destiné, à maintenir la Navarre. La position de Logrono est trop près, il faut
occuper au moins jusqu’à Tudela pour soumettre la Navarre, et tout ce qui
bougeroit. Dans l’ordre offensif, deux divisions peuvent se porter en marche forcée
sur l’armée.
6º. Il ne faut point faire une guerre timide, ni souffrir aucun rassemblement armé
à deux marches d’aucun corps d’armée. Si l’ennemi s’approche, il ne faut point se
laisser décourager par ce qui s’est passé, se confier dans sa supériorité, marcher
à lui et le battre. L’ennemi prendra lui même probablement une marche très
circonspecte: il y sera réduit du moment qu’il aura eu quelque exemple.
Dans cette situation de choses, toutes les fois qu’on seroit sérieusement attaqué
par l’ennemi, on pourra lui opposer le corps du roi, qui doit toujours être ensemble,
et les deux tiers du corps du maréchal Bessières. Se maréchal doit toujours tenir
un tiers de son corps, à une demi journée, un tiers à une journée du corps du
centre, et un tiers sur la droite, suivant les circonstances, également, un tiers du
corps du général Verdier doit se tenir à la gauche du roi, pour le joindre si cela
étoit nécessaire, de sorte que dans un jour le roi puisse réunir 40 mille hommes.
7º. Il faut débuter par des coups d’éclât, qui rélévent le moral du soldat et qui
fassent comprendre à l’habitant qu’il doit rester tranquille, un des premiers coups
le plus important à porter, et qui seroit utile pour réléver l’opinion et compenser
l’évacuation de Madrid, seroit que la brigade du 14me et 44me qu’on rappelle de
Saragosse, aidée d’une détachement du corps du centre, soumette Soria, le
désarme et le fasse rester tranquille. Attaquer et culbuter tout ce qui se présentera
doit être l’instruction générale, donnée au maréchal Bessières, au maréchal Ney,
et au général Verdier, de sorte qu’à une marche, ou à une marche et demie du
corps François, il n’y ait aucun rassemblement d’insurgés; on est d’opinion que si
l’avant garde du général Castaños s’avance sur l’Aranda et dépasse les
montagnes de Buitrago il faut, avec tout ce qu’on réunira dans un jour, marcher à
lui sans lui donner le tems de s’y établir sérieusement, le culbuter, le jetter au delà
des montagnes, et si l’affaire est décisive, se reporter sur Madrid. L’ennemi doit
essayer de déloger l’armée Françoise de cette position, par trois points, par la
Gallice et l’Estramadure, par la droite d’Aranda, et enfin par les rassemblemens
des provinces d’Arragon, de Valence et autres de Castille. Toutes ces
combinaisons sont difficiles à l’ennemi, et si on dissipe ces rassemblemens à
mesure qu’ils se formeront sur tous les points et qu’on les tienne à distance d’une
ou deux marches du cantonnement François, si alternativement les François
prennent l’offensive, tantôt à leur droite, en renforçant le maréchal Bessières,
pendant que le centre se tiendra dans une bonne position derrière la rivière, et à
l’abri de toute attaque, tantôt au centre avec le corps du roi, les deux tiers du corps
de droite, et un tiers du corps de gauche, l’ennemi sera bientôt obligé à la plus
grande circonspection.
8º. On auroit pu aussi conserver Madrid en renforçant le corps qui s’y trouve, du
14me et 44me de ligne, de la brigade du général Mouton, de celle du général Le
Fevbre, qui en dernier lieu a été renvoyée au marshal Bessières, et enfin du
renfort qu’amène le maréchal Ney. On auroit ainsi renforcé le corps de Madrid de
plus de 14 mille hommes, et il est douteux que l’ennemi eut voulu se mesurer avec
des forces aussi considérables et s’exposer à une perte certaine.
9º. Si de fortes raisons obligoient d’évacuer Aranda, on perdroit l’espoir de
rétablir ses communications avec le Portugal. Dans le cas où un évènement
quelconque porteroit à évacuer le Duero et à se concentrer sur Burgos pour se
réunir là avec le maréchal Bessières, le corps du général Verdier peut
communiquer par l’Ebre, et avoir toujours son mouvement isolé pour maintenir la
Navarre, contenir l’Arragon, tous les rassemblemens de ce côté, et protéger la
route principale.[9] Pendant cet intervalle des renforts journaliers arriveront à
l’armée, jusqu’à ce qu’enfin les divisions de la grande armée qui sont en marche,
soient sur les Pyrénées.
On a recommandé de tous tems le petit fort de Pancorvo. Il est nécessaire de
l’occuper, même quand on ne garderoit pas la ligne de l’Ebre, c’est une vedette
d’autant plus utile qu’elle domine la plaine, et seroit un obstacle si jamais l’ennemi
s’en emparoit.[9]
10º. La troisième position qui se présente à l’armée, c’est la gauche à
Pampelune, et la droite sur Vittoria, maintenant ainsi ses communications avec les
places importantes de St. Sebastien et de Pampelune. Au reste toutes ces notes
peuvent difficilement être de quelque utilité, les évènemens modifient
nécessairement les dispositions, tout dépend d’ailleurs de saisir un moment.
11º. Résumé. Le premier but est de se maintenir à Madrid si cela est possible.
Le second, de maintenir ses communications avec le Portugal en occupant la
ligne du Duero.
Le troisième, de conserver l’Ebre.
Le quatrième de conserver ses communications avec Pampelune et St.
Sebastien afin que la grande armée arrivant, on puisse en peu de tems culbuter et
anéantir tous les révoltés.
Lieut.-Gen. Bertrand.
Rochefort, 6 Août, 1808.
APPENDIX.
 APPENDIX.

No. I.

S E C T I O N 1 . — G E N E R A L S TAT E O F T H E F R E N C H A R M Y

IN THE PENINSULA, EXTRACTED FROM THE IMPERIAL

MUSTER ROLLS.
King Joseph commanding, 1st Oct. 1809.
Present under arms Detached. Absent. Effective. Horses.
Men. Horses. Men. Horses. Hosp. Prison. Men. Cav. Draught.
180,814 28,091 10,407 3,165 46,109 4,124 237,330 23,196 8,060
Deduct for the governments 10,407 3,165
——— ———
Real total 226,927 28,091
——— ———

15th July, 1810.

273,403 52,336 29,462 7,846 47,107 4,915 349,972 41,848 18,334
In march to join
6,121 736 ” ” 636 ” 6,757 736
——— ——— ——— ——— ——— ——— ——— ———
279,524 53,072 29,462 7,846 47,743 4,915 356,729 60,918
——— ——— ——— ——— ——— ——— ——— ———

15th August, 1810.

279,639 52,063 25,340 6,017 46,982 5,995 351,961 41,446 16,634
In march to join 1,957 681 511
——— ——— ———
Total effective in Spain 353,918 42,127 17,145
Troops destined for Spain, quartered on the frontier 16,006 1,447 ”
——— ——— ———
Grand total 369,924 43,574 17,145
——— ——— ———

Note.—By this state it appears that allowance being made for casualties, the
reinforcements for Spain, in consequence of the peace with Austria, were not less
than one hundred and fifty thousand men.

15th Jan. 1811.

Present under arms. Detached. Absent. Effective. Horses.
Men. Horses. Men. Horses. Hosp. Men. Cav. Draught.
295,227 52,462 17,780 4,714 48,831 361,838 41,189 15,987
 15th April, 1811.
276,575 46,990 15,121 2,166 40,079 331,776 37,855 11,301

These states shew a decrease of nearly thirty thousand men in three months.
During this period the siege of Badajos, the retreat of Massena, the battles of the
Gebora, Barrosa, and Fuentes Onoro took place. Hence, if the deaths in hospital
be added to the losses sustained in those operations we shall find that, at the
period of its greatest activity, the Guerilla system was more harassing than
destructive to the French army.
 S E C T I O N 2 . — S TAT E O F T H E A R M Y O F P O R T U G A L .

April, 1810.—Head-quarter Caceres. Massena, Prince of Esling, commanding.

Pris- Effec-
Under arms. Detached. Hosp. Horses.
oners. tive.
Men. Horses. Men. Horses. Men. Men. Men. Cav. Draught.
2d corps d’armée 18,372 4,449 1,119 132 1,628 7 21,126 3,520 1,061
6th Ditto 33,759 10,159 496 110 5,086 349 39,690 3,140 3,129
8th Ditto 28,045 7,070 25 ” 5,976 99 34,145 5,312 1,758
Total active army 80,176 21,678 1,640 242 12,690 455 94,961 15,972 5,948
Imperial guards 17,380 3,800 174 15 733 ” 18,287 2,831 954
Province of St. Ander 13,464 752 276 ” 1,774 377 15,891 15,752 ”
Province of Valladolid 4,509 124 123 ” 859 145 6,136 ” 126
Total under Massena’s
116,029 26,254 2,213 257 16,056 977 135,275 19,555 7,056
command

15th May, 1810.

Etat major et gend’armes 229 241 ” ” ” ” 229 241 ”
2d corps Reynier 16,903 2,921 992 231 1,337 42 19,232 2,186 966
6th do. Ney 28,883 5,421 1,224 964 4,940 357 35,067 2,152 4,233
8th do. Junot 30,782 4,228 7 30 5,642 75 2,643 2,142 2,116
Reserve of cavalry. Montbrun 4,776 4,851 246 189 95 ” 5,117 5,040 11
Total active army 71,573 17,662 2,489 1,414 12,014 474 86,076 11,761 7,315

15th August, 1810.

Etat major, &c 199 222 ” ” 3 ” 202 222 ”
2d corps 16,418 2,894 2,494 397 3,006 ” 21,913 1,969 1,304
6th corps 23,456 2,496 1,865 577 5,541 193 30,862 1,701 1,372
8th corps 18,803 1,959 436 169 4,996 98 24,235 2,016 1,112
Reserve of cavalry 4,146 4,322 1,138 831 157 31 5,441 4,907 246
Artillerie et genie et du siege 2,724 2,969 205 159 409 ” 3,339 108 3,128
Total active army 65,746 15,862 6,139 2,119 14,112 302 85,997 10,815 7,162
6th Government Valladolid.
12,693 3,045 639 20 1,775 641 15,107 2,931 134
Division Serras
Asturias et St. Ander. Bonet 12,913 ” 1,394 15 1,578 107 14,885 434 ”
Total under Massena 91,352 18,907 8,172 2,154 17,465 1,050 115,989 13,746 7,296
9th corps, Drouet Comte
19,144 2,436 24 ” 3,147 ” 22,315 2,436 ”
D’Erlon
General Total 110,496 21,343 8,196 2,154 20,612 1,050 138,304 16,616 7,296
 Army of Portugal, 27th September, 1810. The 9th corps to the 15th October.

The reserve of cavalry, and the artillery of siege to the 1st September only.
Under arms. Detached. Hosp. Effective. Horses.
Men. Horses. Men. Horses. Men. Men. Cav. Draught.
Etat major 192 219 ” ” 4 196 219 ”
2d corps 16,575 2,921 2,397 287 2,214 21,186 1,872 1,336
6th do. 23,224 2,478 1,708 600 5,418 30,350 1,730 1,348
8th do. 18,807 2,958 663 140 4,656 24,126 2,027 1,071
Reserve of cavalry 4,146 4,322 1,138 831 157 5,441 4,907 246
Artilleries of siege 3,022 3,115 206 159 409 3,637 146 3,128
Battalion of march which quitted
” ” 474 16 ” 474 16 ”
Bayonne the 2d of October
Total 65,966 16,013 6,586 2,033 12,858 85,410 10,917 7,129
9th corps 19,062 2,072 413 ” 3,516 22,991 1,755 317
Division Serras 8,586 1,015 269 35 1,750 10,605 1,050 ”
Grand Total 93,614 19,100 7,268 2,068 18,124 119,006 13,722 7,446