Professional Documents
Culture Documents
TSV 3D RF Integration High Resistivity Si Interposer Technology 1St Edition Shenglin Ma Ebook Full Chapter
TSV 3D RF Integration High Resistivity Si Interposer Technology 1St Edition Shenglin Ma Ebook Full Chapter
TSV 3D RF Integration High Resistivity Si Interposer Technology 1St Edition Shenglin Ma Ebook Full Chapter
Series Editors
C. P. Wong, Sheng Liu and Wenhui Zhu
TSV 3D RF INTEGRATION
High Resistivity Si Interposer Technology
SHENGLIN MA
Associate Professor, Department of Mechanical and Electrical Engineering, Xiamen
University, China
YUFENG JIN
Professor, Peking University, China
Elsevier
Radarweg 29, PO Box 211, 1000 AE Amsterdam, Netherlands
The Boulevard, Langford Lane, Kidlington, Oxford OX5 1GB, United Kingdom
50 Hampshire Street, 5th Floor, Cambridge, MA 02139, United States
Copyright © 2022 Chemical Industry Press, Co., Ltd. Published by Elsevier Inc.
No part of this publication may be reproduced or transmitted in any form or by any means, electronic or mechanical,
including photocopying, recording, or any information storage and retrieval system, without permission in writing
from the publisher. Details on how to seek permission, further information about the Publisher’s permissions policies
and our arrangements with organizations such as the Copyright Clearance Center and the Copyright Licensing Agency,
can be found at our website: www.elsevier.com/permissions.
This book and the individual contributions contained in it are protected under copyright by the Publisher (other than as
may be noted herein).
Notices
Knowledge and best practice in this field are constantly changing. As new research and experience broaden our
understanding, changes in research methods, professional practices, or medical treatment may become necessary.
Practitioners and researchers must always rely on their own experience and knowledge in evaluating and using any
information, methods, compounds, or experiments described herein. In using such information or methods they
should be mindful of their own safety and the safety of others, including parties for whom they have a professional
responsibility.
To the fullest extent of the law, neither the Publisher nor the authors, contributors, or editors, assume any liability
for any injury and/or damage to persons or property as a matter of products liability, negligence or otherwise, or
from any use or operation of any methods, products, instructions, or ideas contained in the material herein.
ISBN: 978-0-323-99602-0
v
vi Contents
ix
This page intentionally left blank
Preface by Yufeng Jin
With the rapid development of technologies such as big data, the Internet of Things,
5G communications, smart equipment, and mobile terminals, the development of
advanced integrated packaging technologies with higher speed, smaller size, more func-
tions, and lower costs has become a major technical challenge. CMOS SOC technology
has entered the 3-nm node, and IBM announced the first 2-nm manufacturing technol-
ogy on May 6, 2021. However, as the process enters the nanometer or atomic-level era,
integration becomes more and more difficult, and the investment required is beyond
imagination. New materials, new processes, and new integration architectures are
urgently needed to solve the problem of continuous improvement in integration.
In recent years, three-dimensional integration technology has developed rapidly.
The three-dimensional integration of semiconductor chips, modules, or systems can
greatly increase the integration density and become an effective expansion or extension
of Moore’s law. Among them, TSV three-dimensional integration technology has
outstanding performance. Due to its high integration density and wide application range,
it can achieve large-scale production at multiple levels such as chips and modules, and it
has become a research frontier in industry and academia. This technology has been
widely used in the fields of CMOS image sensors, MEMS devices, RF devices and
modules, LED modules, memory stacked modules, and even CMOS SOC.
Worldwide, Georgia Institute of Technology in the United States, IMEC in
Belgium, Fraunhofer in Germany, IME in Singapore, Samsung in South Korea, TSMC,
ST Microelectronics, and others attach great importance to, and vigorously promote,
TSV integration capabilities, and new technologies and products continue to emerge.
Similar to the development history of many new technologies, China has also made
arrangements in this technical field. The 02 National Major Science and Technology
Project, 973, and other programs have successively funded TSV-related technical
research. China Aerospace Science and Technology Corporation and Huawei are also
paying more and more attention and continuing to invest. However, in general, the
foundations in this field are relatively weak, and the cultivating of relevant talents and
core technologies is slow to progress.
As the earliest explorer of TSV basic technology, Dr. Shenglin Ma has accumulated
rich experience after more than 10 years of painstaking research. He independently com-
pleted the development of the first four-layer chip stacked by the TSV process in China
during his doctoral dissertation stage. After working at Xiamen University, he conducted
much with other superior teams focused on 3D RF integration based on TSV interposer
technology, including low-loss RF TSV interposer and the 5–10 GHz 3D integrated
xi
xii Preface by Yufeng Jin
channelized frequency converter receiver. Some of the results have been applied in
engineering fields such as IPD, TSV three-dimensional radio frequency integration
and microchannel heat dissipation, and 38-GHz TSV interposer based antennas. On
this basis, the author combined the current developments in TSV RF integration tech-
nology and wrote this book: “TSV 3D RF Integration: High Resistivity Si Interposer
Technology”.
As the author’s doctoral tutor, I am very happy to see the publication and write the
preface of this book and hope that its publication will play a positive role in promoting the
development of TSV technology. As a partner of the author for many years of collabo-
rative research, I was fortunate to read and review all the manuscripts in advance. This
book reviews the cutting-edge development of HR-Si interposer technology and intro-
duces details of design and process development of HR-Si RF TSV, the design and pro-
cess of a new 3D interconnection structure based on HR-Si TSV, the verification of
2.5D/3D RF heterogeneous integration applications, and HR-Si interposer integrated
inductor and patch antenna. Furthermore, it also introduces the HR-Si interposer with
embedded microchannels. Most of the content of this book is the result of author’s accu-
mulated experience and results of research and development in recent years. The exper-
imental and theoretical analysis are closely integrated, and the content is detailed
and clear.
I believe that this volume will be of great reference value for scientific researchers,
engineering technicians, and graduate students in related fields.
Yufeng Jin
Peking University
Beijing, China
Preface by Shenglin Ma
Since the invention of the transistor, the integrated circuit (IC) industry has contin-
uously developed in line with Moore’s law, by which the transistor’s feature size has
halved every 18 months. The size reduction also results in a cost reduction per
transistor, realizing a higher level of integration density and higher energy efficiency.
However, as Moore’s law is approaching the physical limit, the sharp increase in cost
has seriously affected the marginal revenue increase brought by scaling down the tran-
sistor’s feature size. To maintain or extend Moore’s law, new types of devices and mate-
rials, including gate-all around (GAA) transistors, carbon nanotube technology, Ge
channels, and III–V compound semiconductors, have been proposed and investigated.
Although these devices and materials have led to an improvement in integration level
and functional complexity, challenges still lie ahead. At the same time, compared to
the transistor in ICs whose feature size is continuously being scaled down, advancements
in the size reduction and integration density of microelectronics devices like analog and
radio frequency (RF) integrated circuits and MEMS/NEMS devices have been lagging
behind, as have the line size and precision of the package substrate and assembly substrate,
which have become the main restriction factors for system overall performance.
With this as background, 3D heterogeneous integration of CMOS with analog/RF
devices and MEMS/NEMS devices, which can improve the comprehensive system per-
formance while elevating the integration level and complexity of functions, has become
a key direction of concern within the industry. TSV 3D integration technology can pro-
vide a three-dimensional electrical interconnect running through the chip substrate and
realize high-density 3D integration through multilayer chip stacking. It is considered an
important enabling technology for transcending Moore’s law and is attracting extensive
attention. At present, TSV 3D integration technology has been commercialized in var-
ious fields, including MEMS, CMOS imaging sensors, IR FPA, FPGA, memory ICs, and
SOCs, and an accelerating permeation and expansion trend can be seen. An important
application direction of this technology is the RF field, in the development of next-
generation mobile communication networks and advanced electronic information
systems.
TSV interposer technology was first commercialized in 2011 by the US company
Xilinx in FPGA devices, and its technical and commercial value are now being widely
researched and recognized. Its application in the RF field originated in 2008, when the
Institute of Microelectronics in Singapore proposed the concept of TSV interposer-based
3D RF integration technology. Currently, whether in the advanced electronics informa-
tion system field or the mobile communications field, system-level integrated packaging
xiii
xiv Preface by Shenglin Ma
communicate with researchers who have interests or will be engaged in the related tech-
nology areas. Constructive suggestions or criticisms are appreciated, in hopes of putting
forward the advances in this field.
This book is intended for design engineers, processing engineers, and application
engineers as well as for postgraduate students majoring in advanced electronics packaging.
Shenglin Ma
Xiamen University
National Key Laboratory of Science and Technology
on Micro/Nano Fabrication
Peking University
This page intentionally left blank
Acknowledgments
Thanks to my PhD supervisor, Professor Yufeng Jin from Peking University; coau-
thor of this book: Professors Wei Wang, Min Miao, and Jin Chen; my partners Liulin Hu
and Shuwei He from Chengdu Ganide Technology Corporation Ltd.; my graduate stu-
dents Jun Yan, Han Cai, Yunheng Sun, Mengcheng Wang, Tingting Lian, Yang Yang,
Yanming Xia, Kuili Ren, Luming Chen, Juan Yu, and Nanxin Wang; research assistants
Honglin Gong, Zhizhen Wang, and Lili Wei; and Editor Gang Wu from Chemistry
Industry Press Co., Ltd., China. Without them, it’s unimaginable that this book could
have been completed. I am grateful to them all, and I wish them a bright future.
A special thanks to my friend Professor Daquan Yu, an expert in both industry and
academics, who now is engaged in the commercialization of TGV technology. His excel-
lent work in this field is presented in Chapter 8. Also, thanks go to his team members Tian
Yu and Li Chen, who took part in the drafting of the chapter.
I am very grateful to my partners, my family, and my friends for their unfailing
encouragement and support.
Shenglin Ma, Ph.D., Associated Professor
Xiamen University
National Key Laboratory of Science and Technology on Micro / Nano Fabrication,
Peking University
Beijing, China
xvii
This page intentionally left blank
CHAPTER 1
Introduction to HR-Si
interposer technology
1.1 Background
Since the invention of the transistor, the integrated circuit (IC) industry has continuously
developed by following Moore’s law, by which the transistor’s feature size has been
halved every 18 months. The size reduction also indicates a cost reduction per transistor,
realizing a higher level of integration density and higher energy efficiency. However, as
Moore’s law is approaching the physical limit, the sharp increase in cost has seriously
affected the marginal revenue increase brought by scaling down the transistor’s feature size.
To keep Moore’s law valid, new types of devices and materials, including gate-all around
(GAA) transistors, carbon nanotube technology, Ge channels, and III-V compound semi-
conductors, have been proposed and investigated. Although these devices and materials
have led to an improvement in integration level and functional complexity, challenges still
lie ahead [1–7]. Relative to the transistor in ICs, advancements in the size reduction and
the integration density of other microelectronic devices, such as analog and radio fre-
quency (RF) components and microelectromechanical/nanoelectromechanical system
(MEMS/NEMS) devices, lag behind. The size and the relative precision of the redistri-
bution layer of package substrate and assembly substrate also lag behind. So, the system
overall performance is restricted.
With this background, the CMOS IC has been integrated with analog/RF devices,
MEMS/NEMS devices, and other chip-level 3D devices. This can improve the compre-
hensive system performance while elevating the integration level and complexity of func-
tions [8,9], thereby becoming a key direction within the industry, as is acknowledged
increasingly. Through silicon via (TSV) 3D integration technology can provide the
3D electrical interconnect running through the chip substrate and realize the high-
density 3D interconnects through the multilayer chip stacking. This technology has been
considered an important enabling technology for continuing or transcending Moore’s
law, and has attracted extensive attention. At present, the TSV 3D integration technology
has been commercialized in various fields, such as MEMS, CMOS imaging sensors, infra-
red focal plane arrays, field-programmable grid arrays, memory ICs, and SOCs, and is
demonstrating an accelerated permeation and expansion trend. An important application
direction of this technology is the RF field under the development of next-generation
mobile communication networks and advanced electronic information systems.
in the presence of NMOS and PMOS devices, and the III–V monocrystalline film and
the relative devices need a collaborative design and process development for CMOS
devices. While for the transistor transfer method, the performance of the thin transistor
layer can be easily affected or degraded, it is still being improved in the comprehensive
performance and diversity of integrated devices.
Besides integrating more kinds of transistors on a common substrate, vertical integra-
tion of various high-performance transistors atop an active chip or wafers by stacking
layers of transistors is another popular method. As shown in Fig. 1.1 [10], the III–V sub-
strate wafer, where the fabrication of transistors is finished while the fabrication of the
interconnect of transistors is not completed, is flipped and bonded to the active side
of a CMOS wafer, thinned to expose the buried transistors in the III–V substrate, and
then the fabrication of interconnection between CMOS top of the III–V based transistors
is continued. Fig. 1.1B shows a 250-nm InP heterojunction bipolar transistor (HBT)
stacked on a 130-nm RF CMOS chip developed by HRL Laboratories. In this scheme,
the transistor level alignment is implemented between stacking transistor layers, which
means a high accuracy requirement, and the InP device performance is faced with the
risk of compromise due to integration. In a similar manner, more layers of transistors
can be stacked. One point to be noted is that the stacked layer of transistors may have
finished transistor interconnection; if so, it means that it has changed to chip level inte-
gration and the alignment requirement can be reduced to that between the stacked chips.
In 2019, researchers from Stanford University in the United States demonstrated a sample
of four stacked layers of transistors, including a 1-μm CMOS chip at base, a layer of
carbon nanotube field-effect transistors (CNFET), RRAM, and a CNFET sensing
device layer.
In recent years, advanced packaging technology enabled by TSV, microbump, and
chip or wafer level bonding technology, has inspired new ways to improve the 3D
RF integration level in future, and has offered a chance for a performance breakthrough.
For example, with the microbump bonding process, the active surfaces of GaN devices,
InP-based (HBT), and GaAs devices can be directly bonded on a CMOS chip in the form
of chip-to-wafer or wafer-to-wafer. Based on this technology, BAE and Teledyne Sci-
entific Company have demonstrated an ultrawide band (UWB) ADC sample, as shown
in Fig. 1.2 [11]. The frequency range of UWB is 2.75–8.75 GHz and 14.25–20.25 GHz,
with the most advanced signal-to-noise ratio (over 30 dB) and distortion ratio, demon-
strating a revolutionary breakthrough of performance improvement. By introducing
TSV to the RF microelecronic chip, 3D RF integration can be easily realized by chip
stacking with microbump. Researchers from Teledyne Scientific Company in the
United States have demonstrated a stacked chip module with a 250-nm InP HBT stand-
ing upon a 150-nm GaN HEMT chip. Fig. 1.3 shows a 3D heterogeneous integrated
phased assay module financed by DARPA [12] in the United States, where a glass inter-
poser containing the patch antenna array is stacked on an SiGe beamformer with TSV
(A)
(B)
Fig. 1.1 3D heterogeneous integration by stacking transistor layers developed by US HRL Laboratories: (A) process flow and (B) sample picture.
Introduction to HR-Si interposer technology 5
Fig. 1.2 Microphotograph of COSMOS ADC consisting of InP chip and Si CMOS substrate chip.
interconnect (Fig. 1.3A); the cross-section of 3D integrated sample is shown in Fig. 1.3B,
and the function test results of amplitude control and phase shift are shown in Fig. 1.3C
and D, respectively.
At the assembly level, a 3D multichip module based on advanced ceramic
substrate has been developed, in which a double-sided chip assembly on substrate and
substrate stacking with soldering are used. To further miniaturize the module, an
HR-Si substrate based on a typical MEMS process is proposed to replace the advanced
ceramic substrate or even the relative metal housing, to achieve a full Si packaged RF
module with small form factor. To realize the electrical connection from inside the Si
package to the outside, various methods have been studied, including formation of
big through Si holes with a KOH or TMAH etching process to expose the buried
input/output electrodes, electromagnetic coupling structure, metallized through-hole
by a conformal deposited metal layer on the sidewall of holes, and the introduction of
TSV interconnect in recent years. Fig. 1.4 shows a diagram of the HR-Si package-based
3D RF integration module. Some research institutes claim that an on-shelf product based
on this technology is available.
Whether transistor level or chip level, collaborative design and process development
is needed between the laminated chips in any 3D heterogeneous integration scheme as
stated previously, indicating great constraints must be confronted. The transistor level is
of a high integration level, but its current maturity level is low due to the process of the
ultrathin device layer. Considering the alignment precision requirements and known
good die yield loss, this level presents great technical difficulties and high risk. The chip
level is preferred due to the moderate process requirement and technical advantages in
terms of form factor, integration level, and performance. Nowadays, the HR-Si based
6 TSV 3D RF integration
(A)
(B)
Fig. 1.3 Concept map of (A) 3D integrated phased array cross section, (B) subarray component
developed by DARPA
(Continued)
Introduction to HR-Si interposer technology 7
(C)
(D)
(A)
(B)
(C)
Fig. 1.5 Embedded HR-Si wafer-level heterogeneous integrated packaging technology.
devices at the same time. The 3D vertical integration can be further implemented by TSV
and microbump [13].
In addition, more studies have been done with the HR-Si interposer being taken as an
independent part in 2.5/3D RF integration in recent years, such as in research from
French CEA-Leti, Silex, IMEC, Peking University, Xiamen University, and the
55th, 13th, and 29th Research Institutes of China Electronics Technology Group
Introduction to HR-Si interposer technology 9
Fig. 1.6 The HR-Si interposer-based 2.5D integration concept proposed by CEA-Leti.
10 TSV 3D RF integration
(A)
(B) (C)
Fig. 1.7 Optical microphotographs during the module assembly process of HR-TSV interposer.
Fig. 1.8 Test results (A) Simulated and measured reflection coefficients of antenna. (B) Simulated and
measured gains (dBi) of antenna (H and E planes).
Another random document with
no related content on Scribd:
imagination; she gives way to her first emotions without reflection;
she is caught by things present, and the pleasure she is enjoying
always seems the highest. It has been remarked that she took
pleasure everywhere, not through that indolence of mind that
attaches us to the place where we are, to avoid the trouble of
changing, but by the vivacity of her character which gave her up
entirely to the pleasures of the moment. Paris does not charm her so
much as to prevent her liking the country, and no one of that age has
spoken about nature better than this woman of fashion who was so
much at ease in drawing-rooms, and seemed made for them. She
escapes to Livry the first fine day to enjoy “the triumph of May,” to
“the nightingale, the cuckoo, and the warbler that begin the spring in
the woods.” But Livry is still too fashionable, she must have a more
complete solitude, and she cheerfully retires under her great trees in
Brittany. This time her Paris friends think she will be wearied to
death, having no news to repeat or fine wits to converse with. But she
has taken some serious moral treatise by Nicole with her; she has
found among those neglected books whose last refuge, like that of old
furniture, is the country, some romance of her young days which she
reads again secretly, and in which she is astonished still to find
pleasure. She chats with her tenants, and just as Cicero preferred the
society of the country people to that of the provincial fashionables,
she likes better to talk with her gardener Pilois than with “several
who have preserved the title of esquire in the parliament of Rennes.”
She walks in her Mall, in those solitary alleys where the trees covered
with fine-sounding mottoes almost seem as though they were
speaking to each other; she finds, in fact, so much pleasure in her
desert that she cannot make up her mind to leave it; nevertheless no
woman likes Paris better. Once back there she surrenders herself
wholly to the pleasures of fashionable life. Her letters are full of it.
She takes impressions so readily that we might almost tell in
perusing them what books she has just been reading, at what
conversations she has been present, what drawing-rooms she has
just left. When she repeats so pleasantly to her daughter the gossip of
the court we perceive that she has just been conversing with the
graceful and witty Madame de Coulanges, who has repeated it to her.
When she speaks so touchingly of Turenne she has just left the Hôtel
de Bouillon, where the prince’s family are lamenting his broken
fortunes as well as his death. She lectures, she sermonizes herself
with Nicole, but not for long. Let her son come in and tell her some
of those gay adventures of which he has been the hero or the victim,
she recounts boldly the most risky tales on condition of saying a little
later, “Pardon us, Monsieur Nicole!” When she has been visiting La
Rochefoucauld everything turns to morality; she draws lessons from
everything, everywhere she sees some image of life and of the human
heart, even in the viper broth that they are going to give Madame de
la Fayette who is ill. Is not this viper, which though opened and
skinned still writhes, like our old passions? “What do we not do to
them? We treat them with insult, harshness, cruelty, disdain; we
wrangle, lament, and storm, and yet they move. We cannot overcome
them. We think, when we have plucked out their heart, that they are
done with and we shall hear no more of them. But no; they are
always alive, they are always moving.” This ease with which she
receives impressions, and which causes her to adopt so quickly the
sentiments of the people she visits, makes her also feel the shock of
the great events she looks on at. The style of her letters rises when
she narrates them, and, like Cicero, she becomes eloquent
unconsciously. Whatever admiration the greatness of the thoughts
and the liveliness of expression in that fine piece of Cicero upon
Caesar that I quoted just now may cause me, I am still more touched,
I admit, by the letter of Madame de Sévigné on the death of Louvois,
and I find more boldness and brilliancy in that terrible dialogue
which she imagines between the minister who demands pardon and
God who refuses it.
These are admirable qualities, but they bring with them certain
disadvantages. Such hasty impressions are often rather fleeting.
When people are carried away by a too vivid imagination, they do not
take time to reflect before speaking, and run the risk of often having
to change their opinion. Thus Madame de Sévigné has contradicted
herself more than once. But being only a woman of fashion, her
inconsistency has not much weight, and we do not look on it as a
crime. What does it matter to us that her opinions on Fléchier and
Mascaron have varied, that after having unreservedly admired the
Princesse de Clèves when she read it alone, she hastened to find a
thousand faults in it when her cousin Bussy condemned it? But
Cicero is a politician, and he is expected to be more serious. We
demand that his opinions should have more coherency; now, this is
precisely what the liveliness of his imagination least permits. He
never boasted of being consistent. When he judges events or men he
sometimes passes without scruple, in a few days, from one extreme
to the other. In a letter of the end of October Cato is called an
excellent friend (amicissimus), and the way in which he has acted is
declared to be satisfactory; at the beginning of November he is
accused of having been shamefully malevolent in the same affair,[25]
because Cicero seldom judges but by his impressions, and in a
mobile spirit like his, very different but equally vivid impressions
follow each other very quickly.
Another danger, and one still greater, of this excess of imagination
which cannot control itself is that it may give us the lowest and most
false opinion of those who yield to it. Perfect characters are only
found in novels. Good and evil are so intermingled in our nature that
the one is seldom found without the other. The strongest characters
have their weaknesses, and the finest actions do not spring only from
the most honourable motives. Our best affections are not entirely
exempt from selfishness; doubts and wrongful suspicions sometimes
trouble the firmest friendships, and it may happen at certain
moments that cupidity and jealousy, of which one is ashamed the
next day, flit rapidly through the mind of the most honourable
persons. The prudent and clever carefully conceal all those feelings
which cannot bear the light; those whose quick impressions carry
them away, like Cicero, speak out, and they are very much blamed.
The spoken or written word gives more strength and permanence to
these fugitive thoughts; they were only flashes; they are fixed and
accentuated by writing; they acquire a clearness, a relief and
importance that they had not in reality. Those momentary
weaknesses, those ridiculous suspicions which spring from wounded
self-esteem, those short bursts of anger, quieted as soon as reflected
on, those unjust thoughts that vexation produces, those ambitious
fits that reason hastens to disavow, never perish when once they
have been confided to a friend. One of these days a prying
commentator will study these too unreserved disclosures, and will
use them, to draw a portrait of the indiscrete person who made them,
to frighten posterity. He will prove by exact and irrefutable
quotations that he was a bad citizen and a bad friend, that he loved
neither his country nor his family, that he was jealous of honest
people, and that he betrayed all parties. It is not so, however, and a
wise man will not be deceived by the artifice of misleading
quotations. Such a man well knows that we must not take these
impetuous people literally or give too much credence to what they
say. We must save them from themselves, refuse to listen to them
when they are led astray by passion, and especially must we
distinguish their real and lasting feelings from all those
exaggerations which are merely passing. For these reasons every one
is not fitted to thoroughly understand these letters, every one cannot
read them as they should be read. I mistrust those learned men who,
without any acquaintance with men or experience of life, pretend to
judge Cicero from his correspondence. Most frequently they judge
him ill. They search for the expression of his thought in that
commonplace politeness which society demands, and which no more
binds those who use it than it deceives those who accept it. Those
concessions that must be made if we wish to live together they call
cowardly compromises. They see manifest contradictions in those
different shades a man gives to his opinions, according to the
persons he is talking with. They triumph over the imprudence of
certain admissions, or the fatuity of certain praises, because they do
not perceive the fine irony that tempers them. To appreciate all these
shades, to give things their real importance, to be a good judge of the
drift of those phrases which are said with half a smile, and do not
always mean what they seem to say, requires more acquaintance with
life than one usually gets in a German university. If I must say what I
think, I would rather trust a man of the world than a scholar in this
matter, for a delicate appreciation.
Cicero is not the only person whom this correspondence shows us.
It is full of curious details about all those who had friendly or
business relations with him. They were the most illustrious persons
of the time, and they played the chief parts in the revolution that put
an end to the Roman Republic. No one deserves to be studied more
than they. It must be remarked here, that one of Cicero’s failings has
greatly benefited posterity. If it were a question of some one else, of
Cato for instance, how many people’s letters would be missing in this
correspondence! The virtuous alone would find a place in it, and
Heaven knows their number was not then very great. But, happily,
Cicero was much more tractable, and did not bring Cato’s rigorous
scruples into the choice of his friends. A sort of good-nature made
him accessible to people of every opinion; his vanity made him seek
praise everywhere. He had dealings with all parties, a great fault in a
politician, for which the shrewd people of his time have bitterly
reproached him, but a fault that we profit by; hence it happens that
all parties are represented in his correspondence. This obliging
humour sometimes brought him into contact with people whose
opinions were the most opposite to his, and he found himself at
certain times in close relations with the worst citizens whom he has
at other times lashed with his invectives. Letters that he had received
from Antony, Dolabella, and Curio still remain, and these letters are
full of expressions of respect and friendship. If the correspondence
went further back we should probably have some of Catiline’s, and,
frankly, I regret the want of them; for if we wish to judge of the state
of a society as of the constitution of a man, it is not enough to
examine the sound parts, we must handle and probe to the bottom
the unsound parts. Thus, all the important men of that time,
whatever their conduct may have been, or to whatever party they
may have belonged, had dealings with Cicero. Memorials of all are
found in his correspondence. A few of their letters still exist, and we
have a large number of those that Cicero wrote to them. The private
details he gives us about them, what he tells us of their opinions,
their habits, and character, allows us to enter freely into their life.
Thanks to him, all those persons indistinctly depicted by history
resume their original appearance; he seems to bring them nearer to
us and to make us acquainted with them; and when we have read his
correspondence we can say that we have just visited the whole
Roman society of his time.
The end we have in view in this book is to study closely a few of
these personages, especially those who were most involved in the
great political events of that period. But before beginning this study
it is necessary to make a firm resolution not to bring to it
considerations which belong to our own time. It is too much the
custom now-a-days to seek arms for our present struggles in the
history of the past. Smart allusions and ingenious parallels are most
successful. Perhaps Roman antiquity is so much in fashion only
because it gives political parties a convenient and less dangerous
battle-field where, under ancient costumes, present-day passions
may struggle. If the names of Caesar, Pompey, Cato, and Brutus are
quoted on all occasions, these great men must not be too proud of
the honour. The curiosity they excite is not altogether disinterested,
and when they are spoken of it is almost always to point an epigram
or set off a flattery. I wish to avoid this mistake. These illustrious
dead seem to me to deserve something better than to serve as
instruments in the quarrels that divide us, and I have sufficient
respect for their memory and their repose not to drag them into the
arena of our every-day disputes. It should never be forgotten that it is
an outrage to history to subject it to the changing interest of parties,
and that it should be, according to the fine expression of Thucydides,
a work made for eternity.
These precautions being taken, let us penetrate with Cicero’s
letters into the Roman society of that great period, and let us begin
by studying him who offers himself so gracefully to do us the
honours.
CICERO IN PUBLIC AND PRIVATE LIFE
I
CICERO’S PUBLIC LIFE
Cicero’s public life is usually severely judged by the historians of
our time. He pays the penalty of his moderation. As this period is
only studied now with political intentions, a man like him who tried
to avoid extremes fully satisfies nobody. All parties agree in attacking
him; on all sides he is laughed at or insulted. The fanatical partisans
of Brutus accuse him of timidity, the warmest friends of Caesar call
him a fool. It is in England and amongst us[26] that he has been least
abused, and that classical traditions have been more respected than
elsewhere; the learned still persist in their old habits and their old
admirations, and in the midst of so many convulsions criticism at
least has remained conservative. Perhaps also the indulgence shown
to Cicero in both countries comes from the experience they have of
political life. When a man has lived in the practice of affairs and in
the midst of the working of parties, he can better understand the
sacrifices that the necessities of the moment, the interest of his
friends and the safety of his cause may demand of a statesman, but
he who only judges his conduct by inflexible theories thought out in
solitude and not submitted to the test of experience becomes more
severe towards him. This, no doubt, is the reason why the German
scholars use him so roughly. With the exception of M. Abeken,[27]
who treats him humanely, they are without pity. Drumann[28]
especially overlooks nothing. He has scrutinized his works and his
life with the minuteness and sagacity of a lawyer seeking the grounds
of a lawsuit. He has laid bare all his correspondence in a spirit of
conscientious malevolence. He has courageously resisted the charm
of those confidential disclosures which makes us admire the writer
and love the man in spite of his weaknesses, and by opposing to each
other detached fragments of his letters and discourses he has
succeeded in drawing up a formal indictment, in which nothing is
omitted and which almost fills a volume. M. Mommsen[29] is scarcely
more gentle, he is only less long. Taking a general view of things he
does not lose himself in the details. In two of those compact pages
full of facts, such as he knows how to write, he has found means to
heap on Cicero more insults than Drumann’s whole volume contains.
We see particularly that this pretended statesman was only an egotist
and a short-sighted politician, and that this great writer is only made
up of a newspaper novelist and a special-pleader. Here we perceive
the same pen that has just written down Cato a Don Quixote and
Pompey a corporal. As in his studies of the past he always has the
present in his mind, one would say that he looks for the squireens of
Prussia in the Roman aristocracy, and that in Caesar he salutes in
advance that popular despot whose firm hand can alone give unity to
Germany.
How much truth is there in these fierce attacks? What confidence
can we place in this boldness of revolutionary criticism? What
judgment must we pronounce on Cicero’s political conduct? The
study of the facts will teach us.
I.
II.