Full download test bank at ebookmeta.

com

The Science of Science 1st Edition Dashun Wang

For dowload this book click LINK or Button below

https://ebookmeta.com/product/the-science-of-
science-1st-edition-dashun-wang/

OR CLICK BUTTON

DOWLOAD EBOOK

Download More ebooks from https://ebookmeta.com

More products digital (pdf, epub, mobi) instant
download maybe you interests ...

The Science of Science 1st Edition Wang

https://ebookmeta.com/product/the-science-of-science-1st-edition-
wang/

Philosophy of Science An Introduction to the Central

Issues 1st Edition Wang Wei

https://ebookmeta.com/product/philosophy-of-science-an-
introduction-to-the-central-issues-1st-edition-wang-wei/

Computational Methods and GIS Applications in Social

Science Fahui Wang

https://ebookmeta.com/product/computational-methods-and-gis-
applications-in-social-science-fahui-wang/

Formal Methods in Computer Science Textbooks in

Mathematics 1st Edition Jiacun Wang

https://ebookmeta.com/product/formal-methods-in-computer-science-
textbooks-in-mathematics-1st-edition-jiacun-wang/
The Science of Citizen Science Katrin Vohland (Editor)

https://ebookmeta.com/product/the-science-of-citizen-science-
katrin-vohland-editor/

Artificial Intelligence for Materials Science Springer

Series in Materials Science 312 Yuan Cheng Editor Tian
Wang Editor Gang Zhang Editor

https://ebookmeta.com/product/artificial-intelligence-for-
materials-science-springer-series-in-materials-science-312-yuan-
cheng-editor-tian-wang-editor-gang-zhang-editor/

Computer Science Principles The Foundational Concepts

of Computer Science Kevin Hare

https://ebookmeta.com/product/computer-science-principles-the-
foundational-concepts-of-computer-science-kevin-hare/

The Silences of Science Gaps and Pauses in the

Communication of Science 1st Edition Felicity Mellor
(Editor)

https://ebookmeta.com/product/the-silences-of-science-gaps-and-
pauses-in-the-communication-of-science-1st-edition-felicity-
mellor-editor/

The Science of Waste 1st Edition Spellman

https://ebookmeta.com/product/the-science-of-waste-1st-edition-
spellman/
The Science of Science

This is the first comprehensive overview of the “science of science,” an

emerging interdisciplinary field that relies on big data to unveil the repro-
ducible patterns that govern individual scientific careers and the workings
of science. It explores the roots of scientific impact, the role of productivity
and creativity, when and what kind of collaborations are effective, the
impact of failure and success in a scientific career, and what metrics can
tell us about the fundamental workings of science. The book relies on data
to draw actionable insights, which can be applied by individuals to further
their career or decision makers to enhance the role of science in society.
With anecdotes and detailed, easy-to-follow explanations of the research,
this book is accessible to all scientists and graduate students, policymakers,
and administrators with an interest in the wider scientific enterprise.

Dashun Wang is an Associate Professor at Kellogg School of Management,

and (by courtesy) McCormick School of Engineering, Northwestern Uni-
versity, where he directs the Center for Science of Science and Innovation.
He is a recipient of the AFOSR Young Investigator Award and was named
Poets&Quants Best 40 Under 40 Business School Professors in 2019.

Albert-László Barabási is Robert Gray Dodge Professor of Network

Science and Distinguished University Professor at Northeastern University,
where he directs the Center for Complex Network Research. He holds
appointments in the Departments of Physics and the College of Computer
and Information Science, in the Department of Medicine at Harvard Med-
ical School, at Brigham and Women’s Hospital in the Channing Division of
Network Science, and the Department of Network and Data Science at
Central European University in Budapest, Hungary.
The Science of Science
Dashun Wang
Northwestern University, Evanston, Illinois

Albert-László Barabási
Northeastern University, Boston
University Printing House, Cambridge cb2 8bs, United Kingdom

One Liberty Plaza, 20th Floor, New York, ny 10006, USA

477 Williamstown Road, Port Melbourne, vic 3207, Australia

314–321, 3rd Floor, Plot 3, Splendor Forum, Jasola District Centre,

New Delhi – 110025, India

79 Anson Road, #06–04/06, Singapore 079906

Cambridge University Press is part of the University of Cambridge.

It furthers the University’s mission by disseminating knowledge in the pursuit of

education, learning, and research at the highest international levels of excellence.
www.cambridge.org
Information on this title: www.cambridge.org/9781108492669
doi: 10.1017/9781108610834

© Cambridge University Press 2020

This publication is in copyright. Subject to statutory exception

and to the provisions of relevant collective licensing agreements,
no reproduction of any part may take place without the written
permission of Cambridge University Press.

First published 2021

Printed in Singapore by Markono Print Media Pte Ltd

A catalogue record for this publication is available from the British Library.

isbn 978-1-108-49266-9 Hardback

isbn 978-1-108-71695-6 Paperback

Cambridge University Press has no responsibility for the persistence or accuracy

of URLs for external or third-party internet websites referred to in this publication
and does not guarantee that any content on such websites is, or will remain,
accurate or appropriate.
To Tian
The Science of Science
 CONTENTS
Acknowledgements viii

Introduction 1

Part I The Science of Career 5

1 Productivity of a Scientist 7
2 The h-Index 17
3 The Matthew Effect 28
4 Age and Scientific Achievement 39
5 Random Impact Rule 51
6 The Q-Factor 60
7 Hot Streaks 71

Part II The Science of Collaboration 81

8 The Increasing Dominance of Teams in Science 85
9 The Invisible College 95
10 Coauthorship Networks 102
11 Team Assembly 110
12 Small and Large Teams 124
13 Scientific Credit 134
14 Credit Allocation 147

Part III The Science of Impact 159

15 Big Science 161
16 Citation Disparity 174
17 High-Impact Papers 184
18 Scientific Impact 189
19 The Time Dimension of Science 197
20 Ultimate Impact 209

Part IV Outlook 221

21 Can Science Be Accelerated? 223
22 Artificial Intelligence 231
23 Bias and Causality in Science 241
Last Thought: All the Science of Science 252

Appendix A1 Modeling Team Assembly 254

Appendix A2 Modeling Citations 257
References 270
Index 296
 ACKNOWLEDGEMENTS

As teams are the fundamental engine for today’s innovation, we

must thank our lab members first and foremost. Indeed, our lab
members contributed to this project in some of the most fundamental
ways – not only did their work compose a large part of the initial book
proposal, it also contributed fresh insights which grew into several new
chapters, broadening and deepening our understanding of the subjects
we set out to explore. These team members include Roberta Sinatra,
Chaoming Song, Lu Liu, Yian Yin, Yang Wang, Ching Jin, Hua-Wei
Shen, Pierre Deville, Michael Szell, Tao Jia, Lingfei Wu, Zhongyang He,
Jichao Li, Binglu Wang, Suman Kalyan Maity, Woo Seong Jo, Jian Gao,
Nima Dehmamy, Yi Bu, David Moser, Alex Gates, Junming Huang,
Qing Ke, and Xindi Wang. Our science of science journey has been
infinitely more fun because we had the opportunity to work and learn in
such inspiring company.
Chapter 9 on the invisible college discusses the “bright ambi-
ence” of brilliant minds, a concept that we experienced vividly while
writing this book. Indeed, so many friends and colleagues have been
generous with their time and expertise along the way, that listing some
of those risks excluding many others. We feel compelled to express our
sincere thanks to Luis Amaral, Sinan Aral, Pierre Azoulay, Federico
Battiston, Jeanne Brett, Elizabeth Caley, Manuel Cebrian, Damon
Centola, Jillian Chown, Noshir Contractor, Ying Ding, Yuxiao Dong,
Tina Eliassi-Rad, Eli Finkel, Santo Fortunato, Morgan Frank, Lee Giles,
Danny Goroff, Shlomo Havlin, Cesar Hidalgo, Travis Hoppe, Ian B.
Hutchins, Ben Jones, Brayden King, Rebecca Meseroll, Sam Molyneux,
ix / Acknowledgements

Karim Lakhani, David Lazer, Jessie Li, Zhen Lei, Jess Love, Stasa
Milojevic, Federico Musciotto, Willie Ocasio, Sandy Pentland, Alex
Petersen, Filippo Radicchi, Iyad Rahwan, Lauren Rivera, Matt Salga-
nik, George Santangelo, Iulia Georgescu, Ned Smith, Paula Stephan,
Toby Stuart, Boleslaw Szymanski, Arnout van de Rijt, Alessandro
Vespignani, John Walsh, Ludo Waltman, Kuansan Wang, Ting Wang,
Adam Waytz, Klaus Weber, Stefan Wuchty, Yu Xie, Hyejin Youn.
Among the many people who made this journey possible, our
special thanks go to two of them in particular, whose “bright ambi-
ence” is especially omnipresent throughout the pages and over the
years. Brian Uzzi is not only always generous with his time, he also
has remarkable insight. Time and again he manages to take ideas we’re
struggling to present and effortlessly elevates them. We are also grateful
to Brian for championing the somewhat radical idea that physicists can
contribute to the social sciences. Our colleague James Evans has been a
close friend and collaborator, and several ideas discussed in this book
would have not been born without him, including but not limited to the
concluding remarks on engaging “all the science of science” for the
future development of the field.
We are extremely grateful for the generous research support we
have received over the years. In particular Riq Parra from AFOSR has
been a true believer from the very beginning, when few people knew
what we meant by the “science of science.” Many concepts discussed in
this book would not have been possible without his strong and con-
tinued support. Dashun also wishes to express his special thanks to the
Kellogg School of Management, an institution that offered a level of
support and trust that most researchers only dream of.
Many unsung heroes contributed to this text, collectively log-
ging countless hours to help guide the book along. We have benefited
tremendously from the excellent and dedicated editorial assistance of
Carrie Braman, Jake Smith, and James Stanfill, as well as Enikő Jankó,
Hannah Kiefer, Alanna Lazarowich, Sheri Gilbert, Michelle Guo, and
Krisztina Eleki. Special thanks to Alice Grishchenko who took on the
brave redesign effort, diligently redrawing all the figures and giving a
visual identity to the book. Yian Yin and Lu Liu have been our go-to
people behind the scenes, springing into action whenever we needed
help, and contributing to a variety of essential tasks ranging from data
analysis to managing references.
x / Acknowledgements

Special thanks to our very professional publishing team at

Cambridge, from our enthusiastic champion, Simon Capelin, to the
editorial team who worked tirelessly to move the project forward and
across the finish line: Roisin Munnelly, Nicholas Gibbons, Henry Cock-
burn, and Marion Moffatt. Thank you for all your help, and for putting
up with numerous missed deadlines.
And finally, Dashun would like to thank his wife Tian Shen,
and dedicate this book to her. Thank you, for everything.
 INTRODUCTION

Scientific revolutions are often driven by the invention of new

instruments – the microscope, the telescope, genome sequencing – each
of which have radically changed our ability to sense, measure, and
reason about the world. The latest instrument at our disposal?
A windfall of digital data that traces the entirety of the scientific enter-
prise, helping us capture its inner workings at a remarkable level of
detail and scale. Indeed, scientists today produce millions of research
articles, preprints, grant proposals, and patents each year, leaving
detailed fingerprints of the work we admire and how they come about.
Access to this data is catalyzing the emergence of a new multidisciplin-
ary field, called science of science, which, by helping us to understand in
a quantitative fashion the evolution of science, has the potential to
unlock enormous scientific, technological, and educational value.
The increasing availability of all this data has created an unpre-
cedented opportunity to explore scientific production and reward. Parallel
developments in data science, network science, and artificial intelligence
offer us powerful tools and techniques to make sense of these millions of
data points. Together, they tell a complex yet insightful story about how
scientific careers unfold, how collaborations contribute to discovery, and
how scientific progress emerges through a combination of multiple inter-
connected factors. These opportunities – and the challenges that come
with them – have fueled the emergence of a new multidisciplinary com-
munity of scientists that are united by their goals of understanding
science. These practitioners of science of science use the scientific methods
to study themselves, examine projects that work as well as those that fail,
2 / The Science of Science

quantify the patterns that characterize discovery and invention, and offer
lessons to improve science as a whole. In this book, we aim to introduce
this burgeoning field – its rich historical context, exciting recent develop-
ments, and promising future applications.
We had three core audiences in mind as we wrote this book.
The primary audience includes any scientist or student curious about
the mechanisms that govern our passion, science. One of the founding
fathers of the science of science, Thomas Kuhn, a physicist turned
philosopher, triggered worldwide interest in the study of science back
to 1962 with the publication of The Structure of Scientific Revolutions.
Kuhn’s notion of “paradigm shift” today is used in almost every cre-
ative activity, and continues to dominate the way we think about the
emergence and acceptance of new ideas in science. In many ways, the
science of science represents the next major milestone in this line of
thinking, addressing a series of questions that are dear to the heart of
every scientist but may well lay outside of the Kuhnian worldview:
When do scientists do their best work? What is the life cycle of scientific
creativity? Are there signals for when a scientific hit will occur in a
career? Which kinds of collaboration triumph and which are destined to
for disaster? How can young researchers maximize their odds of suc-
cess? For any working scientist, this book can be a tool, providing data-
driven insight into the inner workings of science, and helping them
navigate the institutional and scholarly landscape in order to better
their career.
A broader impact of the science of science lies in its implications
for policy. Hence, this book may be beneficial to academic adminis-
trators, who can use science of science to inform evidence-based deci-
sion-making. From department chairs to deans to vice presidents of
research, university administrators face important personnel and invest-
ment decisions as they try to implement and direct strategic research.
While they are often aware of a profusion of empirical evidence on this
subject, they lack cohesive summaries that would allow them to extract
signals from potential noise. As such, this book may offer the know-
ledge and the data to help them better take advantage of useful insights
the science of science community has to offer. What does an h-index of
25 tell us about a physics faculty member seeking tenure? What would
the department most benefit from: a junior vs. a senior hire? When
should we invest in hiring a superstar, and what can we expect their
impact will be?
3 / Introduction

We also hope that program directors with National Science

Foundation (NSF), National Institutes of Health (NIH), and other public
and private funding agencies will find the book useful for supporting
high-performing individuals and teams to best address science’s emerging
challenges. Many civilian and military government agencies, nonprofits,
and private foundations are already collecting data and developing tools
rooted in science of science. The framework offered in the coming chap-
ters will allow them to utilize this data in a way that best serves their own
purposes, helping them set up more effective funding mechanisms, and
ultimately benefitting both science and society.
The changing landscape of science also affects scholarly pub-
lishers, who often compete to publish articles that will impact the
direction and the rate of scientific progress. We hope journal editors
will also find science of science useful for a range of practical purposes –
from understanding the natural life cycle of a discovery’s impact to
identifying hit ideas before they become hits – which may, in turn,
augment the impact of what they publish.
Lastly, this book is intended for scientists who are currently
involved in science of science research, or for those who wish to enter
this exciting field. It is our aim to offer the first coherent overview of the
key ideas that currently capture the discipline’s practitioners. Such an
overview is necessary, we believe, precisely because our community is
highly interdisciplinary. Indeed, key advances in the science of science
have been generated by researchers in fields ranging from the infor-
mation and library sciences to the social, physical, and biological sci-
ences to engineering and design. As such, approaches and perspectives
vary, and researchers often publish their results in venues with non-
overlapping readership. Consequently, research on the science of sci-
ence can be fragmented, often along disciplinary boundaries. Such
boundaries encourage jargon, parochial terms, and local values. In the
book we aim to summarize and translate the insights from highly
diverse disciplines, presenting them to students and researchers cohe-
sively and comprehensively. We will not only emphasize the common
intellectual heritage of the diverse set of ideas that coexist in the field,
but also provide approaches to orient new research. Thus, we hope the
book will be a resource for interested students and researchers just
discovering the field.
The book is structured in four parts: The Science of Career
focuses on the career path of individual scientists, asking when we do
4 / The Science of Science

our best work and what distinguishes us from one another. The Science
of Collaboration explores the advantages and pitfalls of teamwork,
from how to assemble a successful team to who gets the credit for the
team’s work. The Science of Impact explores the fundamental dynamics
underlying scientific ideas and their impacts. The Outlook part sum-
marizes some of the hottest frontiers, from the role of AI to bias and
causality. Each part begins with its own introduction which illuminates
the main theme using questions and anecdotes. These questions are then
addressed in separate chapters that cover the science relevant to each.
By analyzing large-scale data on the prevailing production and
reward systems in science, and identifying universal and domain-specific
patterns, science of science not only offers novel insights into the nature
of our discipline, it also has the potential to meaningfully improve our
work. With a deeper understanding of the precursors of impactful
science, it will be possible to develop systems and policies that more
reliably improve the odds of success for each scientist and science
investment, thus enhancing the prospects of science as a whole.
 Part
I THE SCIENCE OF CAREER

Albert Einstein published 248 papers in his lifetime, Charles

Darwin 119, Louis Pasteur 172, Michael Faraday 161, Siméon Denis
Poisson 158, and Sigmund Freud 330 [1]. Contrast these numbers with
the body of work of Peter Higgs, who had published only 25 papers by
the age of 84, when he received the Nobel Prize for predicting the Higgs
boson. Or think of Gregor Mendel, who secured an enduring legacy
with only seven scientific publications to his name [2].
These differences show that in the long run what matters to a
career is not productivity, but impact. Indeed, there are remarkable
differences among the impact of the publications. Even for star scien-
tists, of all papers they publish, at most a few may be remembered by a
later generation of scientists. Indeed, we tend to associate Einstein’s
name with relativity and Marie Curie with radioactivity, while lacking
general awareness of the many other discoveries made by each. In other
words, one or at most a few discoveries – the outliers – seem to be what
define a scientist’s career. So, do these outliers accurately represent a
scientific career? Or did these superstar scientists just get lucky in one or
a few occasions along their careers?
And, if only one or at most a few papers are remembered, when
do scientists make that defining discovery? Einstein once quipped, “A
person who has not made his great contribution to science before the
age of 30 will never do so” [3]. Indeed, Einstein was merely 26 years old
when he published his Annus Mirabilis papers. Yet, his observation
about the link between youth and discovery was not merely autobio-
graphical. Many of the physicists of his generation too made their
6 / The Science of Science

defining discoveries very early in their career – Heisenberg and Dirac at

24; Pauli, Fermi, and Wigner at 25; Rutherford and Bohr at 28. But is
youth a necessity for making an outstanding contribution to science?
Clearly not. Alexander Fleming was 47 when he discovered penicillin.
Luc Montagnier was 51 when he discovered HIV. And John Fenn was
67 when he first began to pursue the research that would later win him
the Nobel Prize in chemistry. So, how is creativity, as captured by
scientific breakthroughs, distributed across the lifespan of a career?
The first part of this book will dive into these sets of fascinating
questions regarding scientific careers. Indeed, as we survey our young
and not so young colleagues doing groundbreaking work, we are
prompted to ask: Are there quantitative patterns underlying when
breakthrough work happens in a scientific career? What mechanisms
drive the productivity and impact of a scientist? The chapters in this part
will provide quantitative answers to these questions, offering insights
that affect both the way we train scientists and the way we acknowledge
and reward scientific excellence.
 1 PRODUCTIVITY OF A SCIENTIST

Paul Erdős, arguably the most prolific mathematician in the

twentieth century, was, by all accounts, rather eccentric. The
Hungarian-born mathematician – who moved to the US before the start
of WWII – lived out of a ragged suitcase that he famously dragged with
him to scientific conferences, universities, and the homes of colleagues
all over the world. He would show up unannounced on a colleague’s
doorstep, proclaim gleefully, “My mind is open.” He then spent a few
days working with his host, before moving on to surprise some other
colleague at some other university. His meandering was so constant that
it eventually earned him undue attention from the FBI. To his fellow
mathematicians, he was an eccentric but lovable scientist. But to law
enforcement officers during the Cold War, it was suspicious that he
crossed the Iron Curtain with such ease. Indeed, Erdős was once
arrested in 1941 for poking around a secret radio tower. “You see,
I was thinking about mathematical theorems,” he explained to the
authorities in his thick Hungarian accent. It took decades of tracking
for the Bureau to finally believe him, concluding that his rambling was
indeed just for the sake of math.
His whole life was, too. He had no wife, no children, no
job, not even a home to tie him down. He earned enough in guest
lecturer stipends from universities and from various mathematics
awards to fund his travels and basic needs. He meticulously
avoided any commitment that might stand in the way of his work.
Before he died in 1996 at the age of 83, Erdős had written or
coauthored a stunning 1,475 academic papers in collaboration
8 / The Science of Science

with 511 colleagues. If total publication counts as a measure of

productivity, how does Erdős’ number compare to the productivity
of an ordinary scientist? It surely seems exceptional. But how
exceptional?

1.1 How Much Do We Publish?

Scholarly publications are the primary mode of communication
in science, helping disseminate knowledge. The productivity of a scien-
tist captures the rate at which she adds units of knowledge to the field.
Over the past century, the number of publications has grown exponen-
tially. An important question is whether the growth in our body of
knowledge is simply because there are now more scientists, or because
each scientist produces more on average than their colleagues in
the past.
An analysis of over 53 million authors and close to 90 million
papers published across all branches of science shows that both the
number of papers and scientists grew exponentially over the past
century [4]. Yet, while the former grew slightly faster than the latter
(Fig. 1.1a), meaning that the number of publications per capita has
been decreasing over time, for each scientist, individual productivity
has stayed quite stable over the past century. For example, the
number of papers a scientist produces each year has hovered at
around two for the entire twentieth century (Fig. 1.1b, blue curve),
and has even increased slightly during the past 15 years. As of 2015,
the typical scientist authors or coauthors about 2.5 papers per year.
This growth in individual productivity has its origins in collabor-
ations: Individual productivity is boosted as scientists end up on
many more papers as coauthors (Fig. 1.1b, red curve). In other
words, while in terms of how many scientists it takes to produce a
paper, that number has been trending downwards over the past
century, thanks to collaborative work individual productivity has
increased during the past decade.

1.2 Productivity: Disciplinary Ambiguities

But, when it comes to a scientist’s productivity, it’s not easy to
compare across disciplines. First, each publication may represent a unit
of knowledge, but that unit comes in different sizes. A sociologist may
9 / Productivity of a Scientist

(a) Overall (b) Average

10 7 5
#papers #authors/paper
#authors #papers/author

Average number
4
Overall number

10 6
3
10 5
2
10 4 1

10 3 0
1900 1925 1950 1975 2000 1900 1925 1950 1975 2000
Year Year
Figure 1.1 The growing number of scientists. (a) During the past century, both
the number of scientists and the number of papers has increased at an exponential
rate. (b) The number of papers coauthored by each scientist has been hovering
around two during the past 100 years, and increased gradually in the past 15 years.
This growth is a direct consequence of collaborative effects: Individual productivity
is boosted as scientists end up on many more papers as coauthors. Similar trends
were reported using data within a single field [5]. For physics, for example, the
number of papers coauthored by each physicist has been less than one during the
past 100 years, but increased sharply in the past 15 years. After Dong et al. [4] and
Sinatra et al. [5].

not feel their theory is fully articulated unless the introduction of the
paper spans a dozen pages. Meanwhile, a paper published in Physical
Review Letters, one of the most respected physics journals, has a strict
four-page limit, including figures, tables, and references. Also, when we
talk about individual productivity, we tend to count publications in
scientific journals. But in some branches of the social sciences and
humanities, books are the primary form of scholarship. While each
book is counted as one unit of publication, that unit is admittedly much
more time-consuming to produce.
And then there is computer science (CS). As one of the youngest
scientific disciplines (the first CS department was formed at Purdue
University in 1962), computer science has adopted a rather unique
publication tradition. Due to the rapidly developing nature of the field,
computer scientists choose conference proceedings rather than journals
as their primary venue to communicate their advances. This approach
has served the discipline well, given everything that has been accom-
plished in the field – from the Internet to artificial intelligence – but it
can be quite confusing to those outside the discipline.
10 / The Science of Science

Ignoring the varying publication conventions that characterize

different disciplines can have serious consequences. For example, in
2017, the US News and World Report (US News), which develops
authoritative ranking of colleges, graduate schools, and MBA programs
around the world, published their first ranking of the world’s best
computer science departments. The ranking was so absurd that the
Computing Research Association (CRA) had to put out a special
announcement, calling it “nonsense” and “a grave disservice” to its
readers.
How could an experienced organization specializing in ranking
academic institutions get it so wrong? It turns out that US News
calculated their rankings based on journal publications recorded by
Web of Science, a procedure that served them well in all other discip-
lines. But, by ignoring peer-reviewed papers published in conferences,
the US News rankings were completely divorced from computer scien-
tists’ own perceptions of quality and impact.
The productivity difference across disciplines can be quanti-
fied using data from the National Research Council on research
doctorate programs in the US [6, 7]. Using the average number of
publications by faculty in each department over a five-year period as
a proxy, researchers find that the numbers ranged from 1.2 in
history to 10.5 in chemistry. Even between similar disciplines we
see large productivity differences. For example, within biological
sciences, faculty productivity ranged from 5.1 in ecology to 9.5 in
pharmacy.
Taken together, the data presented so far in this chapter make
at least one message crystal clear: no matter how we measure it, the
productivity of a typical scientist is nowhere near Erdős’. Indeed, his
total – 1,475 papers – implies a staggering two papers per month over
a span of 60 years. By contrast, a study focusing on more than
15 million scientists between 1996 and 2011, found that less than
1 percent of our colleagues managed to publish at least one paper
every year [8]. Hence, only a small fraction of the scientific workforce
can maintain a steady stream of publications. Interestingly, this small
fraction contains the most high-impact researchers. Though they rep-
resent less than 1 percent of all publishing scientists, this stable core
puts out 41.7 percent of all papers, and 87.1 percent of all papers with
more than 1,000 citations. And if a productive scientist’s pace lags, so
does the impact of their contributions. Indeed, the average impact of
11 / Productivity of a Scientist

papers published by a researcher is substantially lower if they skipped

even a single year.
While Erdős is an outlier, his impressive productivity speaks to
the enormous productivity differences among researchers. Why are
there such differences? After all, we all have a 24-hour day to work
with. So how can people like Erdős be so much more productive than
their peers? To answer these questions, we need to visit the legendary
Bell Laboratory in its heyday.

1.3 Productivity: The Difference

The career of William Shockley, the man who brought silicon to
Silicon Valley, was not free of controversies. To be sure, his attempts to
commercialize a new transistor design in the 1950s and 1960s trans-
formed the Valley into the hotbed of electronics. Yet, his troubling
advocacy for eugenics eventually isolated him from his colleagues,
friends, and family. Shockley spent his most productive years at the Bell
Laboratory, where he co-invented the transistor with John Bardeen and
Walter Brattain. That discovery not only won the trio the 1956 Nobel
Prize in Physics, it also began the digital revolution we continue to
experience today.
While managing a research group at Bell Labs, Shockley
became curious [9]: Were there measurable differences in the prod-
uctivity of his fellow researchers? So he gathered statistics on the
publication records of employees in national labs such as Los Alamos
and Brookhaven. Once he charted the numbers, he was surprised by
the outcome: The curve indicated that individual productivity, the
number of papers published by a researcher, N, follows a lognormal
distribution
 
1 ð ln N  μÞ2
PðN Þ ¼ pffiffiffiffiffi exp  : ð1:1Þ
Nσ 2π 2σ 2

Lognormal distributions are fat-tailed, capturing great variations in

productivity. In other words, Shockley learned that most researchers
publish very few papers, whereas a non-negligible fraction of scientists
are orders of magnitude more productive than the average. Evidence for
(1.1) is shown in Fig. 1.2, plotting the distribution of the number of
papers written by all authors listed in INSPECT, together with a log-
normal fit [10].
12 / The Science of Science

7
10

number of authors with publications

105

103

101
all authors
log-norm. distr.

0 1 2 3 4
10 10 10 10 10
Number of publications

Figure 1.2 Productivity distribution. The blue symbols show the

number of papers published by all authors listed in the INSPECT database
of scientific and technical literature, in the period 1969–2004 (over
3 million authors). The red line corresponds to the lognormal fit to the
data (1.1). After Fronczak et al. [10].

Box 1.1 The study of productivity has a long history [9–15]

In 1926, Alfred J. Lotka [11] observed that the number of papers produced
by scientists follows a fat-tailed distribution. In other words, he found that
a small fraction of scientists are responsible for the bulk of scientific
literature. Lotka studied 6,891 authors listed in Chemical Abstracts pub-
lishing between 1907 and 1916, concluding that the number of authors
making N contributions follows a power law

PðN ÞeN α , ð1:2Þ

where the exponent α  2. A power law predicts that productivity has a

long tail, capturing major variations among individuals. Note that it often
requires a large amount of data to reliably distinguish a power law from a
lognormal distribution [9], which Lotka did not have in 1926.

This lognormal distribution of productivity is rather odd, as

Shockley quickly noticed. Indeed, in most competitive arenas, individual
performance metrics almost always follow a narrow distribution. Think
about running. At the Rio Olympics in 2016, Usain Bolt finished the
100-meter final in just 9.81 seconds. Justin Gatlin came in second and
13 / Productivity of a Scientist

Andre De Grasse in third, with running times 9.89 s and 9.91 s, respect-
ively. These numbers are awfully close, reflecting a well-known fact that
performance differences between individuals are typically bounded [16].
Similarly, Tiger Woods, even on his best day, only took down his closest
contenders by a few strokes, and the fastest typist may only type a few
words more per minute than a merely good one. The bounded nature of
performance reminds us that it is difficult, if not impossible, to signifi-
cantly outperform the competition in any domain. Yet, according to
Fig. 1.2, this boundedness does not hold for scientific performance.
Apparently, it is possible to be much better than your competitors when
it comes to churning out papers. Why is that?

1.4 Why So Productive?

Shockley proposed a simple model to explain the lognormal
productivity distribution he observed (Eq. 1.1) [9]. He suggested that in
order to publish a paper, a scientist must juggle multiple factors, like:

F1. Identify a good problem.

F2. Make progress with it.
F3. Recognize a worthwhile result.
F4. Make a decision as to when to stop the research and start writing up
the results.
F5. Write adequately.
F6. Profit constructively from criticism.
F7. Show determination to submit the paper for publication.
F8. Make changes if required by the journal or the referees.

If any of these steps fail, there will be no publication. Let us assume that
the odds of a person clearing hurdle Fi from the list above is pi. Then,
the publication rate of a scientist is proportional to the odds of clearing
each of the subsequent hurdles, that is N ~ p1p2p3p4p5p6p7p8. If each of
these odds are independent random variables, then the multiplicative
nature of the process predicts that P(N) follows a lognormal distribu-
tion of the form (1.1).
To understand where the outliers come from, imagine, that
Scientist A has the same capabilities as Scientist B in all factors, except
that A is twice as good at solving a problem (F2), knowing when to stop
(F4), and determination (F7). As a result, A’s productivity will be eight
times higher than B’s. In other words, for each paper published by
14 / The Science of Science

Scientist B, Scientist A will publish eight. Hence small differences in

scientists’ ability to clear individual hurdles can together lead to large
variations in overall productivity.
Shockley’s model not only explains why productivity follows
lognormal distribution, but it also offers a framework to improve our
own productivity. Indeed, the model reminds us that publishing a paper
does not hinge on a single factor, like having a great idea. Rather, it
requires scientists to excel at multiple factors. When we see someone
who is hyper-productive, we tend to attribute it to a single exceptional
factor. Professor X is really good at coming up with new problems (F1),
or conveying her ideas in writing (F5). The model suggests, however,
that the outliers are unlikely to be explained by a single factor; rather, a
researcher is most productive when she excels across many factors and
fails in none.
The hurdle model indicates that a single weak point can
choke an individual’s productivity, even if he or she has many
strengths. It also tells us that Erdős may have not been as super-
human as we often think he was, or that his productivity might be
attainable with careful honing of various skills. Indeed, if we could
improve at every step of writing a paper, and even if it’s just a tiny bit
in each step, these improvements can combine to exponentially
enhance productivity. Admittedly, this is easier said than done. But
you can use this list to diagnose yourself: What step handicaps your
productivity the most?
The remarkable variations in productivity have implications
for reward. Indeed, Shockley made another key observation: while
the productivity of a scientist is multiplicative, his salary – a form of
reward often tied to performance – is additive. The highest paid
employees earn at best about 50–100 percent more than their peers.
There are many reasons why this is the case – it certainly seems fairer,
and it helps ensure a collaborative environment. Yet, from a paper-
per-dollar perspective, Shockley’s findings raise some interesting ques-
tions about whether the discrepancy between additive salaries and
multiplicative productivities could be exploited. Indeed, an institution
may be better off employing a few star scientists, even if that means
paying them a great deal more than their peers. Shockley’s arguments
are often used as a rationale for why top individuals at research-
intensive institutions are offered much higher salaries and special
15 / Productivity of a Scientist

perks, and why top departments within a university get disproportio-

nately more funding and resources.
To be sure, gauging a career based on publication count
alone grossly misrepresents how science works. Yet, individual prod-
uctivity has been shown to closely correlate with the eminence of a
scientist as well as her perceived contributions to the field. This
pattern was documented by Wayne Dennis, dating back at least to
1954 [1], when he studied 71 members of the US National Academy
of Sciences and eminent European scientists. He found that, almost
without exception, highly productive individuals have also achieved
scientific eminence, as demonstrated by their listing in the Encyclo-
pedia Britannica or in histories of important developments they have
contributed to the sciences. Higher productivity has been shown to
increase the odds of receiving tenure [17], and of securing funding
for future research [18]. At the institutional level, the publication
rates of the faculty are not only a reliable predictor of a program’s
reputation, they also influence the placement of graduates into fac-
ulty jobs [19].
In sum, sustained high productivity is rare, but it correlates with
scientific impact and eminence. Given this evidence, it may appear that
productivity is the key indicator for a meaningful career in science. Yet,
as we show in the following chapters, among the many metrics used to
quantify scientific excellence, productivity is the least predictive. The
reason is simple: While great scientists tend to be very productive, not
all scientists who are productive make long-lasting contributions. In
fact, most of them do not. Multiple paths can lead to achieving high
productivity. For example, lab technicians in certain fields may find
their names on more than a hundred – or sometimes as many as a
thousand – papers. Hence, they appear to be exceptionally prolific
based on their publication counts, but are rarely credited as the intellec-
tual owner of the research. The way people publish is also changing
[20]. Coauthorship is on the rise, as are multiple publications on
the same data. There have also been more discussions about LPUs,
which stands for least publishable unit [20] or the “salami publishing”
approach, which could further contribute to inflated productivity
counts.
So, if productivity is not the defining factor of a successful
career, what is?
16 / The Science of Science

Box 1.2 Name disambiguation

Our ability to accurately track individual productivity relies on our skill to
identify the individual(s) who wrote a paper and all other work that
belongs to that individual [21, 22]. This seemingly simple task represents
a major unsolved problem [21–23], limited by four challenges. First, a
single individual may appear in print under multiple names because of
orthographic and spelling variants, misspellings, name changes due to
marriage, religious conversion, gender reassignment, or the use of pen
names. Second, some common names can be shared by multiple individ-
uals. Third, the necessary metadata is often incomplete or missing. This
includes cases where publishers and bibliographic databases failed to
record authors’ first names, their geographical locations, or other identify-
ing information. Fourth, an increasing percentage of papers is not only
multi-authored, but also represents multidisciplinary and multi-
institutional efforts. In such cases, disambiguating some of the authors
does not necessarily help assign the remaining authors.
While multiple efforts are underway to solve the name disambigu-
ation problem, we need to be somewhat mindful about the results pre-
sented in this and following chapters, as some conclusions may be affected
by the limitations in disambiguation. In general, it is easier to disambigu-
ate productive scientists, who have a long track record of papers, com-
pared with those who have authored only a few publications. Therefore,
many studies focus on highly productive scientists with unusually long
careers instead of “normal” scientists.
 2 THE h-INDEX

Lev Landau, a giant of Russian physics, kept a handwritten list in

his notebook, ranking physicists on a logarithmic scale of achievement
and grading them into “leagues” [24]. According to Landau, Isaac
Newton and Albert Einstein belonged to the highest rank, above anyone
else: he gave Newton the rank 0 and Einstein a 0.5. The first league, a
rank of 1, contains the founding fathers of quantum mechanics, scientists
like Niels Bohr, Werner Heisenberg, Paul Dirac, and Erwin Schrödinger.
Landau originally gave himself a modest 2.5, which he eventually ele-
vated to 2 after discovering superfluidity, an achievement for which he
was awarded the Nobel Prize. Landau’s classification system wasn’t
limited to famous scientists, but included everyday physicists, who are
given a rank of 5. In his 1988 talk “My Life with Landau: Homage of a 4
1/2 to a 2,” David Mermin, who coauthored the legendary textbook
Solid State Physics, rated himself a “struggling 4.5” [25].
When scientists leave league 5 behind and start approaching the
likes of Landau and other founders of a discipline, it’s obvious that their
research has impact and relevance. Yet for the rest of us, things are
somewhat blurry. How do we quantify the cumulative impact of an
individual’s research output? The challenge we face in answering this
question is rooted in the fact that an individual’s scientific performance
is not just about how many papers one publishes, but a convolution of
productivity and impact, requiring us to balance the two aspects in a
judicious manner.
Of the many metrics developed to evaluate and compare scien-
tists, one stands out in its frequency of use: the h-index, proposed by
18 / The Science of Science

Jorge E. Hirsch in 2005 [26]. What is the h-index, and how to calculate
it? Why is it so effective in gauging scientific careers? Does it predict the
future productivity and impact of a scientist? What are its limitations?
And how do we overcome these limitations? Answering these questions
is the aim of this chapter.

2.1 The h-Index: Definitions and Implications

The index of a scientist is h if h of her papers have at least h
citations and each of the remaining papers have less than h citations
[26]. For example, if a scientist has an h-index of 20 (h = 20), it means
that she has 20 papers with more than 20 citations, and the rest of her
papers all have less than 20 citations. To measure h, we sort an individ-
ual’s publications based on her citations, going from the most cited
paper to the least cited ones. We can plot them on a figure, that shows
the number of citations of each paper, resulting in a monotonically
decreasing curve. Fig. 2.1 uses the careers of Albert Einstein and Peter
Higgs as case studies showing how to calculate their h-index.
Is an h-index of 8, for example, impressive or modest? What is
the expected h-index of a scientist? To answer these questions, let’s take
a look at a simple but insightful model proposed by Hirsch [26].
Imagine that a researcher publishes n papers each year. Let us also
assume that each paper earns c new citations every year. Hence a
paper’s citations increase linearly with its age. This simple model pre-
dicts the scientist’s time dependent h-index as
c
h¼ t: ð2:1Þ
1 þ c=n

Therefore, if we define
1
m  1 þ1 , ð2:2Þ
=c =n

we can rewrite (2.1) as

h = mt, (2.3)

indicating that a scientist’s h-index increases approximately linearly

with time. Obviously, researchers don’t publish exactly the same
number of papers every year (see Chapter 1), and citations to a
paper follow varied temporal trajectories (as we will cover in
19 / The h-Index

(a) Einstein = 67 (b) Higgs =8

104
103
more
than
103 more
citations than
102
102 citations

first papers 101

101
first papers

100 100
100 100 101 100 101
N N
Figure 2.1 The h-index of Albert Einstein (a) and Peter Higgs (b). To calculate
the h-index, we plot the number of citations versus paper number, with papers listed
in order of decreasing citations. The intersection of the 45 line with the curve gives
h. The total number of citations is the area under the curve [26]. According to
Microsoft Academic Graph, Einstein has an h-index of 67, and Higgs 8. The top
three most cited papers by Einstein are: (1) Can quantum mechanical description of
physical reality be considered complete, Physical Review, 1935; (2) Investigations
on the theory of Brownian movement, Annalen der Physik, 1905; and (3) On the
electrodynamics of moving bodies, Annalen der Physik, 1905. The top three for
Higgs are: (1) Broken symmetries and the masses of gauge bosons, Physical Review
Letters, 1964; (2) Broken symmetries, massless particles and gauge fields, Physics
Letters, 1964; (3) Spontaneous symmetry breakdown without massless bosons,
Physical Review, 1966.

Chapter 19). Yet, despite the model’s simplicity, the linear relation-
ship predicted by (2.3) holds up generally well for scientists with
long scientific careers [26].
This linear relationship (2.3) has two important implications:

(1) If a scientist’s h-index increases roughly linearly with time, then its
speed of growth is an important indicator of her eminence. In other
words, the differences between individuals can be characterized by
the slope, m. As (2.2) shows, m is a function of both n and c. So, if a
scientist has higher productivity (a larger n), or if her papers collect
more citations (higher c), she has a higher m. And the higher the m,
the more eminent is the scientist.
(2) Based on typical values of m, the linear relationship (2.3) also
offers a guideline for how a typical career should evolve. For
20 / The Science of Science

example, Hirsch suggested in 2005 that for a physicist at

major research universities, h  12 might be a typical value
for achieving tenure (i.e., the advancement to associate profes-
sor) and that h  18 might put a faculty member into consider-
ation for a full professorship. Fellowship in the American
Physical Society might typically occur around h  15–20, and
membership in the US National Academy of Sciences may
require h  45 or higher.

Since its introduction, the h-index has catalyzed a profusion of

metrics and greatly popularized the idea of using objective indicators to
quantify nebulous notions of scientific quality, impact or prestige [27].
As a testament to its impact, Hirsh’s paper, published in 2005, had been
cited more than 8,000 times as of the beginning of 2019, according to
Google Scholar. It even prompted behavioral changes – some ethically
questionable – with scientists adding self-citations for papers on the
edge of their h-index, in hopes of boosting it [28–30]. Given its preva-
lence, we must ask: can the h-index predict the future impact of a
career?

Box 2.1 The Eddington number

The h-index for scientists is analogous to the Eddington number for
cyclists, named after Sir Arthur Eddington (1882–1944), an English
astronomer, physicist, and mathematician, famous for his work on the
theory of relativity. As a cycling enthusiast, Eddington devised a measure
of a cyclist’s long-distance riding achievements. The Eddington number,
E, is the number of days in your life when you have cycled more than E
miles. Hence an Eddington number of 70 would mean that the person in
question has cycled at least 70 miles a day on 70 occasions. Achieving a
high Eddington number is difficult, since jumping from, say, 70 to 75
may require more than 5 new long-distance rides. That’s because any
rides shorter than 75 miles will no longer be included. Those hoping to
increase their Eddington number are forced to plan ahead. It might be
easy to achieve an E of 15 by doing 15 trips of 15 miles – but turning that
E = 15 into an E = 16 could force a cyclist to start over, since an E number
of 16 only counts trips of 16 miles or more. Arthur Eddington, who
reached an E = 87 by the time he died in 1944, clearly understood that if
he wanted to achieve a high E number, he had to start banking long rides
early on.
21 / The h-Index

2.2 The Predictive Power of the h-Index

To understand the value of the h-index, let’s take a look at the
“usual suspects” – metrics that are commonly used to evaluate a scien-
tist’s performance, and review their strengths and limitations [26].

(1) Total number of publications (N).

Advantage: Measures the productivity of an individual.
Disadvantage: Ignores the impact of papers.
(2) Total number of citations (C).
Advantage: Measures a scientist’s total impact.
Disadvantage: It can be affected by a small number of big hits,
which may not be representative of the individual’s overall
career, especially when these big hits were coauthored with
others. It also gives undue weight to highly cited reviews as
opposed to original research contributions.
(3) Citations per paper (C/N).
Advantage: Allows us to compare scientists of different ages.
Disadvantage: Outcomes can be skewed by highly cited papers.
(4) The number of “significant papers,” with more than c citations.
Advantage: Eliminates the disadvantages of (1), (2), (3), and meas-
ure broad and sustained impact.
Disadvantage: The definition of “significant” introduces an arbi-
trary parameter, which favors some scientists or disfavors others.
(5) The number of citations acquired by each of the q most-cited papers
(for example, q = 5).
Advantage: Overcomes many of the disadvantages discussed above.
Disadvantage: Does not provide a single number to characterize a
given career, making it more difficult to compare scientists to
each other. Further, the choice of q is arbitrary, favoring some
scientists while handicapping others.

The key advantage of the h-index is that it sidesteps all of the

disadvantages of the metrics listed above. But, is it more effective
at gauging the impact of an individual’s work? When it comes to
evaluating the predictive power of metrics, two questions are often
the most relevant.

Q1: Given the value of a metric at a certain time t1, how well does it
predict the value of itself or of another metric at a future time t2?
22 / The Science of Science

This question is especially interesting for hiring decisions. For example,

if one consideration regarding a faculty hire is the likelihood of the
candidate to become a member of the National Academy of Sciences
20 years down the line, then it would be useful to rank the candidates by
their projected cumulative achievement after 20 years. Hirsch tested Q1
by selecting a sample of condensed matter physicists and looked at their
publication records during the first 12 years of their career and in the
subsequent 12 years [31]. More specifically, he calculated four different
metrics for each individual based on their career records in the first 12
years, including the h-index (Fig. 2.2a), the total number of citations

(a) (b)
c(t1)
c(t1)

h(t1) c(t1)

(c) (d)
c(t2)

c(t2)

n(t1) c(t1) / n(t1)

Figure 2.2 Quantifying predictive power of the h-index. Scatter plots compare
the total number of citations, C, after t2 = 24 years vs. the value of the various
indicators at t1 = 12 year for each individual within the sample. Hirsch hypothesized
C may grow quadratically with time, and hence used its square root when calculating
the total number of citations. By calculating the correlation coefficient, he found that
the h-index (a) and the number of citations at t1 (b) are the best predictors of the
future cumulative citations at t2. The number of papers correlates less (c), and the
number of citations per paper performs the worst (d). After Hirsch [31].
23 / The h-Index

(Fig. 2.2b), the total number of publications (Fig. 2.2c), and the average
number of citations per paper (Fig. 2.2d). He then asked if we want to
select candidates that have the most total citations by year 24, which
one of the four indicators gives us the best chance? By measuring the
correlation coefficient between future cumulative citations at time t2 and
four different metrics calculated at time t1, he found that the h-index
and the number of citations at time t1 turn out to be the best predictors
(Fig. 2.2).
While Fig. 2.2 shows that the h-index predicts cumulative
impact, in many cases it’s the future scientific output that matters the
most. For example, if we’re deciding who should get a grant, how many
more citations an applicant’s earlier papers are expected to collect in the
next few years is largely irrelevant. We’re concerned, instead, with
papers that the potential grantee has not yet written and the impact of
those papers. Which brings us to Q2:
Q2: How well do the different metrics predict future scientific output?

To answer Q2, we need to use indicators obtained at t1 to predict

scientific achievement occurring only in the subsequent period,
thereby omitting all citations to work performed prior to t1.
Hirsch repeated the similar prediction task for the four metrics,
but this time used each of them to predict total citations accrued
by papers published only in the next 12 years. Naturally, this is a
more difficult task, but an important one for allocating research
resources. Hirsch found that the h-index again emerges as the
best predictor for achievement incurred purely in future time
frame [31].
These findings indicate that two individuals with similar h are
comparable in terms of their overall scientific achievement, even if their
total number of papers or citations are quite different. Conversely, two
individuals of the same scientific age can have a similar number of total
papers or citation counts but very different h values. In this case, the
researcher with the higher h is typically viewed by the community as the
more accomplished. Together, these results highlight the key strength of
the h-index: When evaluating scientists, it gives an easy but relatively
accurate estimate of an individual’s overall scientific achievements.
Yet at the same time, we must also ask: What are the limitations of
the h-index?
24 / The Science of Science

Box 2.2 The birth of the h-index

Since its inception, the h-index has been an integral part of a scientific life.
Its exceptional influence prompted us to reach out to Jorge Hirsch to ask
how he arrived to the measure. He kindly responded, writing:

I thought about it first in mid 2003, over the next weeks

I computed the h-index of everybody I knew and found that it
usually agreed with the impression I had of the scientist. Shared it
with colleagues in my department, several found it interesting.
Mid June 2005 I wrote up a short draft paper, sent it to
4 colleagues here. One skimmed over it, liked it and made some
suggestions, one liked some of it and was nonplussed by some of
it, two didn’t respond. So I wasn’t sure what to do with it.
Mid July 2005 I got out of the blue an email from Manuel
Cardona in Stuttgart saying he had heard about the index from
Dick Zallen at Virginia Tech who had heard about it from one of
my colleagues at UCSD (didn’t say who but I can guess). At that
point I decided to clean up the draft and post it in arXiv, which
I did August 3, 2005, still was not sure what to do with it.
Quickly got a lot of positive (and some negative) feedback, sent
it to PNAS August 15.

2.3 Limitations of the h-Index

The main street of College Hill in Easton, Pennsylvania – the
home of the Lafayette College – is named after James McKeen Cattell.
As an American psychologist, Cattell played an instrumental role in
establishing psychology as a legitimate science, advocacy that prompted
the New York Times to call him “the dean of American science” in his
obituary.
While many have thought of developing new metrics to system-
ically evaluate their fellow researchers, Cattell was the first to popular-
ize the idea of ranking scientists. He wrote in his 1910 book, American
Men of Science: A Biographical Directory [32]: “It is surely time for
scientific men to apply scientific method to determine the circumstances
that promote or hinder the advancement of science.” So, today’s obses-
sion of measuring impact using increasingly sophisticated yardsticks is
25 / The h-Index

by no means a modern phenomenon. Scientists have been sizing up their

colleagues since the beginning of the discipline itself. A century after
Cattell’s book, the need and the rationale for a reliable toolset to
evaluate scientists has not changed [33].
As the h-index has become a frequently used metric of scientific
achievements, we must be mindful about its limitations. For example,
although a high h is a somewhat reliable indicator of high accomplish-
ment, the converse is not necessarily always true [31]: an author with a
relatively low h can achieve an exceptional scientific impact with a few
seminal papers, such as the case of Peter Higgs (Fig. 2.1b). Conversely, a
scientist with a high h achieved mostly through papers with many
coauthors would be treated overly kindly by his or her h. Furthermore,
there is considerable variation in citation distributions even within a
given subfield, and subfields where large collaborations are typical (e.g.,
high-energy experimental physics) will exhibit larger h values, suggest-
ing that one should think about how to normalize h to more effectively
compare and evaluate different scientists.
Next we discuss a few frequently mentioned limitations of the
h-index, along with variants that can – at least to a certain degree –
remedy them.

 Highly cited papers. The main advantage of the h-index is that its
value is not boosted by a single runaway success. Yet this also means
that it neglects the most impactful work of a researcher. Indeed, once
a paper’s citations get above h, its relative importance becomes invis-
ible to the h-index. And herein lies the problem – not only do outlier
papers frequently define careers, they arguably are what define sci-
ence itself. Many remedies have been proposed to correct for this [34–
39], including the g-index (the highest number g of papers that
together received g2 or more citations [40, 41]) and the o-index (the
geometric mean of the number of citations gleaned pffiffiffiffiffiffiffiffiby a scientist’s
∗
highest cited papers c and her h-index: o = c∗ h [42]). Other
measures proposed to correct this bias include a-index [36, 38];
h(2)-index [39]; hg-index [34]; q2-index [37]; and more [35].
 Inter-field differences. Molecular biologists tend to get cited more
often than physicists who, in turn, are cited more often than math-
ematicians. Hence biologists typically have higher h-index than physi-
cists, and physicists tend to have an h-index that is higher than
mathematicians. To compare scientists across different fields, we
26 / The Science of Science

must account for the field-dependent nature of citations [43]. This can
be achieved by the hg-index, which rescales the rank of each paper n
by the average number of papers written by author in the same year
and discipline, n0 [43] or the hs-index, which normalizes the h-index
by the average h of the authors in the same discipline [44].
 Time dependence. As we discussed in Chapter 2.2, the h-index is time
dependent. When comparing scientists in different career stages, one
can use the m quotient (2.2) [26], or contemporary h-index [45].
 Collaboration effects. Perhaps the greatest shortcoming of the h-index
is its inability to discriminate between authors that have very different
coauthorship patterns [46–48]. Consider two scientists with similar
h indices. The first one is usually the intellectual leader of his/her
papers, mostly coauthored with junior researchers, whereas the second
one is mostly a junior author on papers coauthored with eminent
scientists. Or consider the case where one author always publishes
alone whereas the other one routinely publishes with a large number
of coauthors. As far as the h-index is concerned, all these scientists are
indistinguishable. Several attempts have been proposed to account for
the collaboration effect, including fractionally allocating credit in
multi-authored papers [48–50], and counting different roles played
by each coauthor [51–54] by for example differentiating the first and
last authorships. Hirsch himself has also repeatedly acknowledged this
issue [46, 47], and proposed the hα-index to quantify an individual’s
scientific leadership for their collaborative outcomes [47]. Among all
the papers that contribute to the h-index of a scientist, only those
where he or she was the most senior author (the highest h-index among
all the coauthors) are counted toward the hα-index. This suggests that a
high h-index in conjunction with a high hα/h ratio is a hallmark of
scientific leadership [47].

In addition to these variations of the h-index, there are other metrics to

quantify the overall achievement of individual scientists, including the
i10-index, used exclusively by Google Scholar [55], which computes the
number of articles with at least 10 citations each; or the SARA method
[56], which uses a diffusion algorithm that mimics the spreading of
scientific credits on the citation network to quantify an individual’s
scientific eminence. Despite the multitude of metrics attempting to
correct the shortcomings of the h-index, to date no other bibliometric
27 / The h-Index

index has emerged as preferable to the h-index, cementing the status of

the h-index as a widely used indicator of scientific achievement.
As we dug deeper into h-index and the voluminous body of
work motivated by it, it was easy to forget a perhaps more important
point: No scientist’s career can be summarized by a single number. Any
metric, no matter how good it is at achieving its stated goal, has
limitations that must be recognized before it is used to draw conclusions
about a person’s productivity, the quality of her research, or her scien-
tific impact. More importantly, a scientific career is not just about
discoveries and citations. Rather, scientists are involved in much
broader sets of activities including teaching, mentoring, organizing
scientific meetings, reviewing, and serving on editorial boards, to name
a few. As we encounter more metrics for scientific eminence, it’s import-
ant to keep in mind that, while they may help us understand certain
aspects of scientific output, none of them alone can capture the diverse
contributions scientists make to our community and society [57, 58].
Just as Einstein cautioned: “Many of the things you can count, don’t
count. Many of the things you can’t count, do count.”
Therefore, we must keep in mind that the h-index is merely a
proxy to quantify scientific eminence and achievement. But the problem
is, in science, status truly matters, influencing the perception of quality
and importance of one’s work. That’s what we will focus on in the next
chapter, asking if and when status matters, and by how much.
 3 THE MATTHEW EFFECT

Lord Rayleigh is a giant of physics, with several laws of nature

carrying his name. He is also known beyond the profession thanks to
Rayleigh scattering, which answers the proverbial question, “Why is the
sky blue?” Rayleigh was already a respected scientist when, in 1886, he
submitted a new paper to the British Association for the Advancement
of Science to discuss some paradoxes of electrodynamics. The paper was
promptly rejected on the grounds that it did not meet the journal’s
expectation of relevance and quality. Yet, shortly after the decision,
the editors reversed course. Not because anything changed about the
paper itself. Rather, it turns out that Rayleigh’s name had been inadvert-
ently omitted from the paper when it was first submitted. Once the
editors realized it was Rayleigh’s work, it was immediately accepted
with profuse apologies [59, 60]. In other words, what was initially
viewed as the scribblings of some “paradoxer,” suddenly became worth
publishing once it became clear that it was the work of a world-
renowned scientist.
This anecdote highlights a signaling mechanism critical in sci-
ence: the role of scientific reputation. Robert K. Merton in 1968 [60]
called this the Matthew effect after a verse in the biblical Gospel of
Matthew pertaining to Jesus’ parable of the talents: “For to everyone
who has will more be given, and he will have an abundance. But from
the one who has not, even what he has will be taken away.” The
Matthew effect as a concept has been independently discovered in
multiple disciplines over the last century, and we will encounter it again
in Chapter 17, when we discuss citations. In the context of careers, the
29 / The Matthew Effect

Matthew effect implies that a scientist’s status and reputation alone

can bring additional attention and recognition. This means that status
not only influences the community’s perception of the scientist’s cred-
ibility, playing an important role in how her work is evaluated, but it
also translates into tangible assets – from research funding to access to
outstanding students and collaborators – which in turn further
improve her reputation. The goal of this chapter is to unpack the role
of the Matthew effect in careers. When does it matter? And to what
extent?

3.1 What’s in a Name?

The Internet Engineering Task Force (IETF) is a community of
engineers and computer scientists who develop the protocols that run
the Internet. To ensure quality and functionality, engineers must submit
all new protocols as manuscripts that undergo rigorous peer review. For
a while, each manuscript included the name of every author. However,
beginning in 1999, some manuscripts replaced the full author list with a
generic “et al.,” concealing the name of some authors from the review
committee.
By comparing cases where well-known authors were hidden by
the et al. label with those where the hidden names were little-known,
researchers effectively conducted a real-world Lord Rayleigh experi-
ment [61]. They found that when an eminent name was present on a
submission, like the chair of a working group, which signals profes-
sional standing, the submission was 9.4 percent more likely to be
published. However, the “chair effect” declined by 7.2 percent when
the senior author’s name was masked by the et al. label. In other words,
name-based signaling accounts for roughly 77 percent of the benefits of
having an experienced author as a coauthor on the manuscript.
Interestingly, when the analysis was restricted to a small pool
of manuscripts that were “pre-screened,” or closely scrutinized, the
author name premium disappeared. This suggests that the status effect
only existed when the referees were dealing with high submission rates.
In other words, when the reviewers do actually read the manuscript,
carefully judging their content, status signals tend to disappear.
Given the exponential growth of science, we frequently encoun-
ter the “too many to read” situations. Yet, typically, peer review is a
rather involved process, with multiple rounds of communication
30 / The Science of Science

between authors and expert reviewers, suggesting that the status signal-
ing may be less of a concern for scientific manuscripts. Indeed, through
those rebuttals and revisions, an objective assessment of the work is
expected to prevail. Yet, as we see next, the status effect is rarely
eliminated.
Whether an author’s status affects the perceived quality of his/
her papers has been long debated in the scientific community. To truly
assess the role of status, we need randomized control experiments,
where the same manuscript undergoes two separate reviews, one in
which the author identities are revealed and another in which they are
hidden. For obvious ethical and logistical reasons, such an experiment
is difficult to carry out. Yet, in 2017, a team of researchers at Google
were asked to co-chair the program of the Tenth Association for
Computing Machinery International Conference on Web Search and
Data Mining (WSDM), a highly selective computer science conference
with a 15.6 percent acceptance rate. The researchers decided to use the
assignment as a chance to assess the importance of status for a paper’s
acceptance [62].
There are multiple ways to conduct peer review. The most
common is the “single-blind” review, when the reviewers are fully
aware of the identity of the authors and the institution where they
work, but, the authors of the paper are not privy to the reviewer’s
identity. In contrast, in “double-blind” review, neither the authors nor
the reviewers know each other’s identity. For the 2017 WSDM confer-
ence the reviewers on the program committee were randomly split into a
single-blind and a double-blind group. Each paper was assigned to four
reviewers, two from the single-blind group and two from the double-
blind group. In other words, two groups of referees were asked to
independently judge the same paper, where one group was aware of
who the authors were, while the other was not.
Given the Lord Rayleigh example, the results were not surpris-
ing: Well-known author – defined as having at least three papers
accepted by previous WSDM conferences and at least 100 computer
science papers in total – were 63 percent more likely to have the paper
accepted under single-blind review than in double-blind review. The
papers under review in these two processes were exactly the same,
therefore, the difference in acceptance rate can only be explained by
author identity. Similarly, authors from top universities had a 58 percent
increase in acceptance once their affiliation was known. Further, for
Another random document with
no related content on Scribd:
The Project Gutenberg eBook of Aatetoverit
This ebook is for the use of anyone anywhere in the United States
and most other parts of the world at no cost and with almost no
restrictions whatsoever. You may copy it, give it away or re-use it
under the terms of the Project Gutenberg License included with this
ebook or online at www.gutenberg.org. If you are not located in the
United States, you will have to check the laws of the country where
you are located before using this eBook.

Title: Aatetoverit
Sosiaalinen romaani

Author: Max Kretzer

Translator: Lauri Soini

Release date: November 18, 2023 [eBook #72161]

Language: Finnish

Original publication: Hämeenlinna: Arvi A. Karisto Oy, 1909

Credits: Juhani Kärkkäinen and Tapio Riikonen

*** START OF THE PROJECT GUTENBERG EBOOK

AATETOVERIT ***
AATETOVERIT

Sosiaalinen romaani

Kirj.

MAX KRETZER

Suomentanut

Juho Ahava

Hämeenlinnassa, Arvi A. Karisto, 1909

Niin kauvan kuin vielä on kaksi ihmistä, joiden

ulkonainen asema on erilainen, niin kauvan tulevat myös
Kain ja Abel olemaan.
I

"Gustav, sinä jäät —"

"Ja minä sanon sinulle, Wilhelm, en ikipäivinä! Sinä olet päässyt

hyvinvointiin, olet rakentanut oman pesäsi, mutta minä olen iäti
jäänyt retkaleeksi. Silloin on parempi, että vetäydyn tieheni,
ennenkuin minulle käy ero vaikeammaksi kuin tänään."

"Ja mihin tahdot mennä, mikä on päämääräsi?"

Puhuteltu vaikeni tuokion, vaivuttaen katseensa lattiaan. Hänen

rinnastaan pääsi huokaus, sitte vastasi hän hiljaa:

"Jumala sen tiennee, en minä. Kenties sinne missä minunlaiseni

ovat: halveksitut ja kurjat, joilla ei ole vaimoa eikä lapsia, ei isää eikä
äitiä, ei sisaria eikä veljiä, joilla ei ole kotia, ei isänmaata eivätkä saa
mistään tointa. Ja kaikki vain siksi, että heillä on sydäntä rinnassaan
köyhyyden kärsimyksiä kohtaan ja he tahtovat nälkäisille hankkia
riittävän leivän. Sentähden, jää hyvästi — en saa häiritä elämäsi
rauhaa."

Puhuja koetti irrottaa kätensä toisen kädestä, mutta tämä ei sitä

hellittänyt, vaan sanoi uudelleen:
 "Ei, Gustav, sanon vielä kerran, sinä jäät. Uskotko tosiaankin
minusta sellaista, että päästäisin sinut, parhaan entisen
koulutoverini, muitta mutkitta menemään, nyt, kun vuosien jälkeen
onnellinen sattuma on vienyt meidät yhteen — juuri nyt, kun minulla
jotakin on eikä sinulla mitään? Oletko sitte unhottanut, että me
ajamme samaa aatetta, pyrimme samaan päämäärään, ihmisten
yleisen veljestymisen suureen päämäärään? Oletko unhottanut, että
me jo koulussa jaoimme sämpylät keskenämme? — Ei, siitä ei tule
mitään, sanon vielä kolmannen kerran: sinä jäät. Kun kerran
puolueemme on kirjottanut lippuunsa, ettei kellään ihmisellä ole
oikeutta ylellisyyteen niin kauvan kuin joltakin toiselta puuttuu
välttämättöminkin, niin jokainen tämän lipun tunnustaja on
velvollinen sen mukaan toimimaan. Alottakaamme siis me. Kun
vaimoni keittää kahdelle, niin keittää hän kai kolmannellekin. Kun
meillä on makuupaikka kahdelle, löytyy kai tilaa sinullekin. Kun
meillä on riittänyt pitää aina hyvä juhlapuku tarpeen varalta, niin
tuottaa liike kyllä vielä siksi, että voidaan pitää huolta myös vanhan,
jälleen löydetyn ystäväni ulkoasusta. Seis siis — kättä päälle vaan,
ja menkäämme takaisin taloon: loppu selviää kyllä siellä sisällä.
Luulen, että vaimoseni on lopettanut työnsä puutarhassa — oi, häntä
en ole vielä lainkaan ajatellut, ja sinä tahtoisit mennä pois häntä
näkemättä, oppimatta häntä tuntemaan! — Sitä en ikinä antaisi
sinulle anteeksi. Niin, ystävä, minulla on oikea mallivaimo, enkä voi
sinulle sanoa, kuinka onnellinen olen ja kuinka tyytyväisinä toisiimme
elämme. Tule siis vastustelematta, pienessä lehtimajassa on niin
mukava jutella pienen välipalan ja kylmän ryypyn ääressä."

Kaksi miestä, jotka näin keskustelivat, seisoivat kiviportailla

yksikerroksisen, siistiltä näyttävän pikkukaupungin talon oven
edessä. Kuuman kesäkuun iltapäivän kirkas päiväpaiste lepäsi
kadulla, jolla tuskin näkyi yhtään ihmistä. Jokainen arasteli
kuumuutta, ja jollei hänen välttämättä täytynyt mennä ulos, tunsi hän
itsensä onnelliseksi viileässä asunnossaan. Mihin silmäsikään
vastapäätä sijaitsevaan taloriviin, joka lepäsi täydessä
auringonvalossa, olivat ikkunaverhot ja käärekaihtimet alaslasketut,
ja harvojen kauppapuotien akkunatelttaimet kaartuivat alas niin
kauvas seinästä kuin mahdollista. Sieltä näkyi sekatavarakaupan ja
maustekauppiaan kilpi; niiden rinnalla sikari- ja tupakkakauppiaan:
loitompaa häämötti kangaskaupan näytetilkku, ja seuraavan
poikkikadun kulmassa komeili loitolle loistavana, keskelle
katukäytävää ulottuvalla, huimaavasti koukertelevalla
rautakannattimella jättiläissuuruinen, kullattu leijona, jonka mukaan
sen takana sijaitseva majatalo ja ravintola oli saanut nimensä.

Jos taloja tarkasteli lähemmin, täytyi niiden harmooniseen

kauneuteen ja suurkaupunkien ylellisen elämän mukavuuksiin
tottuneeseen tuntua poroporvarillisen yksinkertaisilta ja kaikkea
edistyneempää rakennustaitoa vailla olevilta. Kaikki näytti
kuivakiskoiselta, säännöttömältä, vinolta ja kulmikkaalta. Sitä
enemmän täytyi pistää silmään rakennuksen, joka, melkein keskellä
kaupunkia sijaiten, kolmine uusimpaan tyyliin rakennettuine
kerroksineen, pani kuin mikäkin jättiläinen toiset näyttämään
kääpiöiltä. Tämän rakennuksen melkein koko pohjakerroksen
ikkunat olivat suurista peililasiruuduista, joiden takaa näkyi
kaikenlaisia huonekaluja. Kaikilla akkunatelttaimilla ja
sisäänkäytävän puoleisella seinällä oli luettavana sama kirjotus:
"Franz Ehlertin huonekalukauppa".

Suoraan tämän talon vastapäätä sijaitsi ensiksi mainittu pieni talo.

Tämänkin valkeaksi kalkitulle seinälle oli oven yläpuolelle pienin
mustin kirjaimin maalattu kirjotus: "Wilhelm Schorn, sorvari ja
puuseppä", saattoi lukea siitä jokainen ohikulkija.
 Tämä talo oli kaikessa tapauksessa muhkein vertaisistaan. Jollei
se ollutkaan suurempi kuin useimmat muut, vaikkei se näistä
lainkaan eronnut rakennustavan ajanmukaisuuteen nähden, niin
täytyi kuitenkin kävelijän katseen ennen kaikkia muita viivähtää siinä
mielihyvällä. Ensimäisen kerroksen neljästä ikkunasta loistivat
puhtaaksi pestyjen peilikirkkaiden ikkunain lävitse kauneimmat
kukat, joiden näkeminen sai jokaisen rauhallisesti sykkivän sydämen
ihastuksesta värähtämään. Ja ikkunain puitteet välkkyivät äskettäin
siveltyinä tummanvihreällä, aivan kuin niiden pitäisi muodostaa
kukille sopusuhtainen kehys.

Ulko-ovi oli maalattu samalla värillä; paksu messinkinen kääkkä

pisti välkkyvänä esiin tummasta pohjasta. Sen oikealla puolella oli
pieni puoti, jonka näyteikkunasta ja kohtuullisen korkeasta lasiovesta
saattoi nähdä, että vasta nyttemmin oli muuan seinä murrettu, jotta
myymälä suurenisi.

Lasioveen oli sisäpuolelle kiinnitetty mitä erilaisimpia

kävelykeppejä. Pienessä näyteikkunassa oli sopusuhtaisesti
järjestettynä kaikellaisia myytäviä esineitä, jollaisia sorvariliikkeessä
tavallisesti on myytävänä: pitkiä ja lyhyitä piippuja kirjavine
nauhoineen ja tupsuineen, samaten valkosia ja ruskeita savipiippuja.
Niiden vieressä oli merenvaha-imukkeita, yksinkertaisesti ja
taitehikkaasti vuoltuja. Ja ikkunan editse juoksevassa nuorassa
riippui porsliinisia piipunpesiä, joihin oli maalattu kauniita nais-,
metsästys- ja muita kuvia. Kaikenlaisia koruesineitä oli riveissä ja
ruoduissa maassa.

Ei tomuhiukkastakaan ollut esineillä havaittavissa, kaikesta tunsi

järjestystä rakastavan ihmisen käden. Katu oli pikemmin kaupungin
laidalla, mutta varsinaisen liikekaupungin asukkaat torin varrella,
kaupunginosassa, jossa suuret, loistavat myymäläholvit sijaitsivat ja
missä varakkaimmilla ja ylhäisimmillä oli tapana tehdä ostoksensa —
eivät edes hekään voineet mennä sorvaripuodin ohitse heittämättä
tarkastavaa, mieltynyttä katsetta näyteikkunaan. Ja milloin ne
tarvitsivat joitakin Wilhelm Shcornin tavaroita, silloin saattoi lyödä
vetoa sadasta yhtä vastaan, etteivät he kammoksuneet kulkea
pitempää taivalta, sillä he tiesivät ostettujen esineiden mukana
tuovansa kotiin tietoisuuden siitä, että olivat kohtuullisella hinnalla
vaihtaneet lujia tavaroita.

Yksi niistä miehistä, joiden keskustelua kuuntelimme, oli Wilhelm

Schorn, mainitun pikkutalon omistaja; toisen nimi oli Gustav
Rassmann.

Schorn oli kolmikymmeninen mies. Soikeahkoilla, hieman

kalpeahkoilla kasvoilla oli säännölliset piirteet; lyhyeksi leikattu
tummanvaalea täysiparta antoi kasvoille todellisen miehekkään
leiman. Vahva, kiharainen tukka peitti pään ja osan korkeata otsaa.
Hänen vartalonsa oli tuskin keskikokoista pitempi, hoikka, vaikkakin
lujarakenteinen; sentähden näytti hän suuremmalta kuin oikeastaan
oli. Hartiat olivat tavallista enemmän koholla, kaula näytti sentähden
sisäänpainuneelta, selkä hieman kumaralta. Eri ammattien tehtäväin
tuntijat, jotka voivat työmiehen ulkonäöstä kohta päättää, mitä hän
tekee, tietävät, että korkeat hartiat johtuvat alituisesta työskentelystä
höylä- ja sahauspenkin ääressä — enimmäkseen höylän
heilutuksesta. Wilhelm Schorn oli paitsi sorvaamista oppinut
muunkin puusepän ammatin ja pikkukaupungissa löytänyt siinä
määrin tilaisuutta sen harjottamiseen, että hänellä oli vakituista työtä
kahdelle kisällille.
 Niinpä oli hänellä naapurilleenkin, huonekalukauppias Franz
Ehlertille milloin sitä milloin tätä tehtävää ja oli sentähden tämän kera
säännöllisissä liikesuhteissa.

Esiintymisessään oli hän hieman avuton, kuten tavallisesti

henkilöt, jotka aina ovat hyväntahtoisia ja ystävällisiä ja sentähden
tuntuvat vähän tyhmiltä.

Vaikka Schorn olisi pukeutunut soreimpaankin muotipukuun, ei

hän olisi voinut konsanaan kieltää ammattiaan.

Sininen esiliina riippui sittenkin hänen edessään, hengessä

näyttäytyi se jokaisen silmiin. Se ei jäänyt pois, vierailipa hän
naapuristossa, seisoi puotitiskinsä ääressä tai kenties iltapäivisin
pistäytyi hetkiseksi "Kultaiseen leijonaan" lukemaan tuoreimpia
pääkaupungin sanomalehtiä.

Aivan toisellainen oli Rassmann ulkonäöltään.

Hän oli itse sanonut ainaiseksi jääneensä retkaleeksi. Ja ottaen

huomioonsa hänen ulkoasunsa tällä kertaa täytyi tunnustaa hänen
olevan oikeassa.

Hän näytti rappeutuneelta ja repaleiselta — täydelleen tuuliajolle

joutuneelta, kuten henkilö, joka on tottunut paremmissa olosuhteissa
elämään ja koettaa vielä kurjuuttansakin pukea eräänlaiseen
loistoon.

Mutta vaikka muodinmukainen pystykauluksensa olisi näyttänyt

vieläkin likaisemmalta, vaikka hänen sinisilkkisessä kravatissaan
olisi ollut tahroja vieläkin enemmän, vaikka hänen kesätakkinsa olisi
näyttänyt vieläkin haalistuneemmalta ja enemmän nukkavierulta,
hänen housunsa ja hänen saappaansa vieläkin kuluneemmilta ja
vahamaisemmilta — olisi häntä nähdessään sentään jokaisen
täytynyt olla sitä mieltä, että hän oli mies, jolla oli sivistystä ja älyä ja
joka tunsi seuraelämän muodot. Erittäinkin miellytti ja kiinnitti hänen
kasvojensa henkevä leima. Kasvoja koristi hyväkasvuinen täysiparta,
piirteet olivat hienot, mutta niitä uursivat hurjan intohimon jäljet;
lisäksi tulivat vaihtelevat kasvojen ilmeet, jotka antoivat hänen
eleilleen jotakin epävakaista.

Jos Rassmannin kasvojen alituisista liikkeistä olisi saanut päättää

jotakin hänen sisäisiin tuntoihinsa nähden, niin olisi tullut siihen
päätökseen, että hänen rinnassaan täytyi käydä erilaisten tunteiden
lakkaamaton vaihtelu, tunteiden, joita hän koetti ulkomaailmalta
salata, mutta joka kuvastui hänen kasvoistaan ja silmistään. Rypyt
hänen kas voissaan, se seikka, että hänen yhä vielä aukoton
tukkansa alkoi harmaantua, tekivät hänet vanhemmaksi kuin hän oli,
sillä hän oli vain kaksi vuotta vanhempi kuin Wilhelm Schorn.

Tehtyään viimeisen vaatimuksensa jälleenlöydetylle ystävälleen,

tarttui Schorn tämän käsivarteen ja koetti vetää häntä puoleksi
hellävaroen, puoleksi väkivetoon mukanaan.

Mutta Rassmann ponnisteli uudelleen vastaan ja silmäili pukuaan.

Sitte sanoi hän:

"Mutta en tiedä, rakas ystävä — sinä saat hävetä minun tähteni; ja

mitä ajatteleekaan rouvasi, kun näkee minut tässä asussa, minut,
jota hän ei lainkaan tunne!"

Wilhelm Schorn nauroi tällöin.

 "Eikö tunne sinua? Jos se on ainoa huolesi, niin silloin ovat
kärsimyksesi lopussa. Vaimoni tietää vallan hyvin, kuka Gustav
Rassmann on, hän vain ei tähän päivään asti ole miestä nähnyt,
vaikka usein on ilmottanut sitä haluavansa. Luuletko sitte, ettemme
lue täällä sanomalehtiä emmekä tiedä, mitä tapahtuu ulkona
suurissa kaupungeissa, varsinkin pääkaupungissa, jossa kaikki
suuret asiat tapahtuvat? Kuinka usein olen kertonutkaan hänelle
sinusta, kun sattumalta luimme nimesi jonkin suuren kokouksen
kertomuksesta. Sinustahan on tullut koko oppinut, kuten sanotaan,
silloin voit kyllä käsittää, kuinka henkeään pidättäen vaimoni
kuunteli, kun kerroin että olet ollut kanssani samassa koulussa, joka
minun valitettavasti täytyi jättää niin aikaisin; ja että me
pääkaupungissa kymmenen vuotta sitte olimme jo parhaat ystävät.
Ylipäänsä ajattelee hän samoin kuin mekin, siis — mutta tuolla hän
jo on, nyt ei auta mikään rimpuileminen."

He olivat jo puutarhan ovella, kun Schorn huusi äänekkäästi:

"Hannaseni, Hannaseni — missä olet?"

Peremmältä puutarhasta tuli oven eteen näkyviin vaalea puku ja

heleä ääni vastasi:

"Täällä, Wilhelm! Tulen jo."

Rouva Schorn aikoi juuri astua ulos ovesta, kun hän paitsi
miestään huomasi vielä vieraankin olevan tulossa luokseen. Se
pidätti hänet, mutta pani hänet uteliaasti silmäilemään valosta
puolipimeään esimajaan.

Ja hänen miehensä huusi edelleen iloisen vallattomasti, vetäen

Rassmanin mukanaan:
 "Hannaseni, Hannaseni, auta minua, muuten muuttaa hän vielä
mielensä ja kääntyy ympäri. Arvaas vain, kenen tuon, minkä rakkaan
muuttolinnun, jota toivottavasti emme toistaiseksi päästä
pesästämme."

Rouva Schorn ei voinut selittää miehensä huimapäisyyttä; hän

vastasi:

"Mutta Wilhelm, kuinka sen voisin tietää."

Nyt tuli näkyviin pari lapsenkasvoja, joista toiset kuuluivat

hoipertelevalle kiharapäiselle kolmivuotiaalle pojalle ja toiset
viisivuotiaalle, mitä suloisimmalle, valkokiharaiselle tytölle. Poika oli
kaikesta päättäen juuri ponnistellut vetääkseen perässään raskaasti
kuormitettuja pieniä santarattaita, mutta oli siinä onnistunut niin, että
nyt makasi maassa pitkin pituuttaan ja osa rattaiden sisällöstä
puolialastomilla säärillään. Hän alkoi surkeasti huutaa, pikku siskon
auttaessa häntä jälleen nousemaan, muttakun hän älysi
pitkäpartaisen vieraan miehen, lakkasi hän heti itkemästä ja hiipi
äitinsä taakse, pitäen pelokkaasti kiinni tämän hameesta; tyttönen
sillävälin katsoi suurin silmin vieraaseen.

Molemmat miehet olivat astuneet puutarhaan.

Pieni mäyräkoira, lasten uskollinen leikkitoveri, joka tähän asti oli

rauhallisesti maannut päiväpaisteessa, hypähti jalkeille kovasti
haukkuen ja yritti hyökätä Rassmannin kimppuun.

"Hiljaa, Ami, tiedä huutia!" sanoi rouva Schorn, mutta elukka, joka
muuten totteli jokaista käskyä ja pian tuli tutuksi, ei tänään
hyökkäysinnoissaan ottanut talttuakseen. Raivossaan lainkaan
asettumatta koetti se tarrata Rassmannin sääreen, ja vasta kun
Schorn pontevasti ärähti: "Koetapas — mars matkaasi!" ja huiskasi
uhkaavasti kädellään, vetäytyi koira syrjään; mutta murina, jolla se
väistymistään säesti, ilmaisi selvään, että se odotti ainoastaan ensi
tilaisuutta hyökkäyksensä uudistaakseen.

Schorn, josta elukan vihaisuus oli hullunkurinen, todennäköisesti

siksi, että hän harvoin näki Amia tällä tuulella, sanoi nauraen
Rassmannille:

"Katsos vain rakkia! Se käyttäytyy aivan kun sinä olisit pahin

vihollisemme etkä kotimme tervetullein vieras."

Sitten kääntyi Schorn vaimoonsa, joka, odottamattomasta

käynnistä yhä hämillään, seisoi syrjässä vierasta katsellen.

"Hannaseni, tässä hän on — kukas muu kuin Gustav Rassmann,

rakas, hyvä Gustavini, hyvän asiamme esitaistelija, jonka ovat
Berlinistä karkoittaneet; tuskin oli tie hänet tuonut kaupunkiimme,
kun hän tahtoi taas lähteä, sellainen veitikka", — hän, nauroi
uudelleen — "aivan kuin sinua ja minua ei täällä olisikaan. Ja
tiedätkös, miksei hän tahtonut viivähtää talossamme? Siksi, että hän
luulee, ettet sinä tunne häntä — sinäkö et häntä tuntisi
sanomalehdistä ja minun kertomuksistani, hahaha — moista
puhetta!"

Schorn näytti aivan hullaantuneelta uudistuneesta ilosta ja hyvästä

tuulesta. Sitte sanoi hän jälleen:

"Niin — ja tässä, Gustav, on rouvaseni, vaimokultaseni, onneni,

kaikkeni. Ja lyökää nyt kättä, tervehtikää toisianne ja tulkaa hyviksi
ystäviksi."
 Samaan hengenvetoon lisäsi hän:

"Mutta olinhan melkein unhoittaa — nämä pienokaiset tässä. Willy,

Lise, tulkaas tänne ja antakaa setä Rassmannille kättä — Willy,
miten muljotatkaan noin, setä Rassmann ei ole mikään musta mies."

Pieni kiharapää, joka yhä vielä piteli kiinni äitinsä hameesta, ei

juuri näyttänyt olevan isänsä kera samaa mieltä, sillä hän alkoi
jälleen itkeä ja tähysteli ympärinsä etsien mäyräkoiraa, ikäänkuin
tämä vastarinnassaan olisi hänen parhain ystävänsä.

Ja ikäänkuin Ami olisi hänet ymmärtänyt, alkoi se uudelleen

haukkua ja tuli parilla hyppäyksellä Rassmannin luo.

Mutta Schorn lopetti poikaansa nähden jutun lyhyeen. Hän otti

tämän syliinsä ja painoi pitkän suutelon pyöreäposkisen pojan suulle.
Se näytti palauttavan Willyn suunnilleen, sillä hän rauhottui ja silmäili
pelottomasti pitkäpartaista miestä sinisillä silmillään. Mutta kun
tämäkin tahtoi antaa hänelle suutelon, ryöstäytyi hän irti ja peitti
kasvonsa isänsä olkapäähän.

Ja Schorn sanoi jälleen nauraen:

"Se pitää sinua myös vihollisena ja pahana vieraana."

Rouva Schornin kasvoilla huomasi ilmeistä iloa Rassmannin

nimeä mainittaessa.

Uteliaisuus hälveni pian sydämellisiin tulotervehdyksiin.

"Mikä yllätys", sanoi hän, laskien kätensä Rassmannin käteen. "Oi,

mieheni on minulle kertonut niin paljo teistä, ettette ole minulle niin
vieras kuin kenties luulette. Olkaa sydämellisesti tervetullut."
 "Siinä nyt näet, mitä jo sanoin", puuttui Schorn samassa
puheeseen. Sitte pyörähti hän ympärinsä kolmivuotias poikansa
käsivarrellaan, kuten mies, joka on erittäin hilpeällä tuulella.

Gustav Rassmann oli kumartanut ystävänsä rouvalle: samassa

muisti hän taasen rappeutuneen ulkoasunsa. Hänet valtasi äkillinen
häpeän ja epämiellyttävyyden tunne.

Kuinka saattoikaan hän tietää, että hänen aatetoverinsa rouva

näytti tuollaiselta!

Rouva Johanna Schorn oli viehättävä rouva. Ainoa vilkaisu riitti

huomatakseen Rassmannista, että hänellä uhraavan toimintansa
ohella ihmisoikeuden puolustamiseksi riitti aina aikaa myöskin
naiskauneuden ihailuun. Ja Berliinissä, hienostuneiden nautintojen ja
hillittömän elämän Babelissa, oli hän tottunut antamaan
intohimoilleen ja itsekkäisyydelleen täyden vapauden — hän, joka
tuomitsi ihmissydämen itsekkäisyyttä.

Kuinka oli Schorn, ulkonäöltään niin yksinkertaisen

poroporvarillinen pikkukaupunkilaisen perikuva, saanut tällaisen
rouvan?

Kun Rassmann sai ensi vaikutuksensa tästä sirosta ja kuitenkin

kukoistavasta valkoverisestä naisesta ja sitten heitti katseen
kulmikkaan ja kömpelön ystävänsä kyyristyneeseen selkään, ei
hänen mieleensä voinut olla heräämättä tämä kysymys.

Ja kuinka sievästi hän oli puettu! Heleä kesäpuku sopi hänelle

moitteettomasti, vyötärö oli solakka ja rinta kuin taltalla veistetty.
Lumivalkea kaulus puki kaunista kaulaa oivallisesti ja väljistä hihoista
helottivat käsivarret kuin rusottavat alabasterit. Kasvot sitten, hieman
aistillinen suu, säännöllisesti muodostunut nenännykerö ja suuret
ruskeat silmät! Tämän odottamattoman näyn edessä joutui
Rassmann äkkiä hämmennyksiin. Hän näki miehen onnen, kuuli
lasten sydämellisen naurun, näki puutarhan kukkasloiston
ympärillään ja hänestä tuntui tosiaan kuin hän olisi pienessä
paratiisissa. Ja hän kuitenkin oli vain halveksittu olento.

Ja jälleen tunsi hän hunningolle joutuneen ulkoasunsa —

selvemmin kuin koskaan tämän naisen vastassa, joka seisoi kuin
tuores ruusu hänen edessään!

Hän sammalsi muutamia sanoja pyytäen anteeksi — ettei hän

"juuri näyttänyt parhaimmalta", ja niin edelleen; Schorn keskeytti
hänet kohta:

"Lörpötystä — kylliksi siitä jo, Gustav! Ikäänkuin vaimoni ei

ajattelisi samoin kuin sinä ja minä. Sinä olet hyvän asiamme
marttyyri, ja kuka sinut on tehnyt siksi? Ei kukaan muu kuin
vastustajamme. Mutta meidän keskuudessamme olkoon
tunnussanana: 'Yksi kaikkien, kaikki yhden puolesta. Mikä minun on,
on myös sinun.’ Ja silleen jääköön. Etkö sinäkin ole sitä mieltä
Hannaseni?"

"Kyllä, Wilhelm."

Vaimo nyökäytti päätänsä, mutta tuli sangen vakavaksi.

Schorn jatkoi;

"Sentähden, Gustav, lyö kättä kanssani ja vielä kerran vaimonikin

kanssa. Ja kuten nyt teemme sinulle pyhän lupauksen, että
pidämme sinua parhaana ystävänämme luonamme parempiin päiviin
saakka, niin tee myös meille mieliksi ja jätä kaikki toimeentulon
huolet toistaiseksi mielestäsi. Aika tuo neuvoja — ja sanon sinulle,
että se päivä on koittava, jolloin meidän aatteemme toteutuvat ja
rehellinen työ pääsee oikeuteensa. Oi, jos voisin, tahtoisin mielelläni
kuolla tehdäkseni kaikki ihmiset onnelliseksi, ja minä puolestani
toivon itselleni vain sen verran, että voin vaimoni ja lasteni kera elää
tyytyväisenä; mitäs siitä vaikkapa minun täytyisi tehdä työtä
enemmänkin kuin nykyisin teen; juuri työstä, näetkö, saan
siunauksen maailmassa. — Kättä siis, eikä sanaakaan enempää."

Schorn oli puhunut jommoisellakin innostuksella; huomasi

jokaisesta sanasta, että niillä tarkoitettiin totta.

Rassmann ojensi hänelle vain äänetönnä kätensä; rouvalle

samoin.

Silloin tuli Schorn jälleen entiselle hauskalle tuulelleen. Hän sanoi:

"Mutta nyt joutuin, Hannaseni, anna tytön huolehtia kahvista,

meidän kai on paras jäädä lehtimajaan. Käske noutaa myös
leivoksia äläkä unhota kisällejä. Tänään on juhlapäivä, sentähden en
enää liikautakaan kättäni. Meillä on yllinkyllin puhuttavaa."

Ja rouva Schorn riensi taloon taluttaen Liisaa kädestä, jollaikaa

nuori mestari, poika yhä käsivarrellaan, istuutui ystävänsä kera
varjoisaan lehtimajaan.

Vielä illan tullen istuivat he siellä ja keskustelivat kaikenlaisista

asioista ja tuumista. Rouva Schorn kuunteli äänetönnä miesten
keskustelua, heittäen vain silloin tällöin sanan väliin. Mutta kun hän
kohotti katseensa, lepäsivät Rassmannin silmät yhä hänen
kasvoillaan.
 Myöhään viedessään ystäväänsä makuulle ei Schorn voinut olla
kysäsemättä:

"No kuinka miellyttää sinua vaimoni, enkö ole hyvin valinnut?"

Ja kun Rassmann nyökkäsi, lisäsi toinen:

"Niin, jokainen sanoo samaa. Ei, sinun on vain päästävä

tietämään kuinka viisas hän on. Hän oli mestarivainajani ainoa tytär
ja tämä talo hänen isänsä tämän eläessä. Olen tosiaan onnellinen ja
tyytyväinen — nuku makeasti, hyvää yötä. Jumala varjelkoon sinua
kattoni alla."