Long live the scientists: Tracking the scientific fame of great minds in physics

Full text

Manuscript Details
Manuscript number JOI_2018_46_R2
Title Long live the scientists: Tracking the scientific fame of great minds in physics
Article type Research Paper
Abstract
This study utilizes global digitalized books and articles to examine the scientific fame of the most influential physicists.
Our research reveals the greatest minds are gone but not forgotten. Their scientific impacts on human history have
persisted for centuries. We also find evidence in support of own-group fame preference, i.e., that the scientists have
greater reputations in their home countries or among scholars sharing the same languages. We argue that, when
applied appropriately, Google Books and Ngram Viewer can serve as promising tools for altmetrics, providing a more
comprehensive picture of the impacts scholars and their achievements have made beyond academia.
Keywords Scientific fame; own-group preference; Google corpus; altmetrics
Manuscript region of origin Asia Pacific
Corresponding Author Li Tang
Corresponding Author's
Institution
Fudan University
Order of Authors Guoyan Wang, Guangyuan Hu, Chuan-Feng Li, Li Tang
Submission Files Included in this PDF
File Name [File Type]
Cover letter-R2.docx [Cover Letter]
updated-20180823-R2-response letter.docx [Response to Reviewers]
20180728-highlights.docx [Highlights]
updated-20180823-long live-without authors.docx [Manuscript File]
To view all the submission files, including those not included in the PDF, click on the manuscript title on your EVISE
Homepage, then click 'Download zip file'.
Research Data Related to this Submission
There are no linked research data sets for this submission. The following reason is given:
Data will be made available on request

1
Long live the scientists: Tracking the scientific fame of great minds in physics
Abstract
This study utilizes global digitalized books and articles to examine the scientific fame of the most
influential physicists. Our research reveals that the greatest minds are gone but not forgotten. Their
scientific impacts on human history have persisted for centuries. We also find evidence in support of
own-group fame preference, i.e., that the scientists have greater reputations in their home countries
or among scholars sharing the same languages. We argue that, when applied appropriately, Google
Books and Ngram Viewer can serve as promising tools for altmetrics, providing a more
comprehensive picture of the impacts scholars and their achievements have made beyond academia.
Key words: Scientific fame; own-group preference; Google corpus; altmetrics
Some say that a man dies three times. The first time is when his heart stops beating and he dies
physically. The second is when people come to his funeral and his identity is erased from society. The
third time is when nobody on the earth remembers him anymore. Then he is really dead.
- “Dragon Raja” by Lee Yeongdo
1. Introduction
Books are the stepping stones to human progress. According to UNESCO (United Nations
Educational, Scientific and Cultural Organization), the number of estimated published books in 2017
alone is up to 2.2 million.1 Such a large collection is undoubtedly a rich archive of human history and
civilization. Yet, as one of the most telling embodiments of knowledge stock and advancement,
books have not captured sufficient attention in quantitative research evaluation.
Fortunately, with access to Google Books and the Google Books tool Ngram Viewer, scholars are
now able to trace cultural evolution on a long time scale based on digitalized texts and trillions of
words. This application of high-throughput data collection to study human culture can be traced back
to Michel et al. (2011). In this pioneering study, the authors utilized the Google Books corpus and
conducted text-based statistical analysis to trace cultural trends. That innovative research method
soon captured academia’s attention and was adopted in the arenas of digital history (Sternfeld, 2011),
the history of science (Laubichler, Maienschein and Renn, 2013), economics (Roth, 2013), social
psychology (Greenfield, 2013; Acerbi, 2013; Zeng and Greenfield, 2015), and cultural psychology
(Pettit, 2016).
Google Books has also been utilized to assess the fame of great scientists throughout history. Based
on the word frequency of people’s full names mentioned in books, Bohannon’s Science Hall of Fame
was built a as an objective evaluation of scientific fame over centuries (Bohannon, 2011a; 2011b).
Moving beyond previous work, this paper utilizes both Google Books, which covers 36 million
global digital books, and Google Scholar, which indexes 91 million academic items,2 to examine the
scientific fame of top physicists. We particularly focus on and compare two of the greatest physicists,
Isaac Newton and Albert Einstein, depicting their fame evolution over centuries and exploring what
1 Data source: http://www.worldometers.info/books/. Accessed on January 18, 2018.
2 The types of academic items Google Scholar includes are research articles, books, patents, case laws, and citations. In
this research our search excludes patents and citations.

2
they are famous for.
Our research reveals that the great minds are gone but not forgotten. Early scientists are still on the
public’s lips in modern society. Their scientific impacts on human history have persisted for
centuries. This holds true for other prominent physicists as well. We also found that while Einstein’s
scientific fame has exceeded that of Newton among intellectuals since the mid-20th century
worldwide, there is a different pattern of fame when the own-group preference is differentiated by
the language of digitalized corpus. The computational analysis confirms that the influence of
Einstein is largely related to his two contributions on general relativity and quantum theory, while
the most frequently mentioned scientific achievements of Newton are the law of universal gravitation
and the laws of motion.
This paper makes the following contributions to the literature. To begin with, this is the first attempt,
within our best knowledge, to explore scientific fame based on the combination of both books and
articles indexed by Google. In addition to depicting the evolution of fame of great minds, we also
explore their most accredited achievements based on co-occurence analysis. Second, our study
contributes to the discussion on the expected role of scientists in science promotion. By comparing
the indicator for scientific impact (i.e., work being cited by scholarly publications) and the indicator
for scientific fame (i.e., name being mentioned in books), our study sheds some light on how to
gauge scientists’ contributions beyond academia. We argue that, when applied appropriately, Google
Books and Google Ngram Viewer can serve as promising tools for altmetrics, providing a more
comprehensive picture of the impacts scholars and their achievements have made on society.
The rest of the paper is structured as follows. In the next section, we delineate our method and case
selection justifications. Section 3 presents our analysis. In Section 4, following a summary of main
findings, we conclude our paper discussing limitations and future research venues.
2. Method and Data
2.1 Notion and measurement
There is no agreed-upon definition of scientific fame or its measurement. The term can be traced
back to the book The Life of Sir Charles Linnæus, a biography of a Swedish botanist, physician, and
zoologist whose fame is centered on his enduring achievement of binomial nomenclature (Stöver
1794). Yet many scientists are ordinary folks and are little known to the public (Astin 1957; Menard
1971; Merton 1970). Some scholars have argued that the fame of scientists should be confined to
professional achievements, while others believe scientific fame goes beyond academia (Menard 1971;
Bohannon 2011a). Feist (2016) noted that regardless of either intrinsic or extrinsic research, the
assessment of fame in science ultimately rests on productivity and its impact on advancing the
research front. Previous studies often used being elected to prestigious societies or winning research
awards or prizes as proxy indicators of scientific recognition or reputation (Bronk, 1976; Youtie, etc.
2013). Instead of relying on the subjective judgment of panel experts, Bohannon’s Science Hall of
Fame (2011a) innovatively uses the appearance of people’s names in books to capture scientists’
influence across different domains throughout history. This is the approach we adopt in tracking and
recording the fame of great scientific minds and their achievements.
2.2 Case selection
The focus domain in this work is physics. Among the myriad scientists, we purposely choose Isaac
Newton and Albert Einstein for illustration based on the following considerations.
To begin with, both Newton and Einstein are in the field of physics, which makes their comparison
relatively free from discipline differences of publication distribution. And given that both names
consist of two words with similar length, the quality of their retrievals is fairly comparative. As the

3
two most influential physicists in the history of science (Whittaker, 1943; Baker, 1984), Newton and
Einstein are appropriate candidates for evaluating the historical fame of individuals. Finally, who is
more influential has been a topic attracting much attention in the global scientific community. The
debate has not been settled for more than half a century (Gribbin, 1987; Graneau and Graneau, 1993).
Successor scientists have commemorated them on special anniversaries, such as “Science 1943:
Aristotle, Newton, Einstein, the three-hundredth anniversary of Newton’s birth” and “New Scientist
1987: Newton vs. Einstein, the three-hundredth anniversary of Newton’s theory of gravity.” In 2005,
the Year of World Physics and also the Centenary of Einstein’s Special Relativity Theory, the UK
Royal Society conducted polls of both academia and the public. The results showed that both Royal
Society members and the British netizens surveyed considered Isaac Newton, a British scientist, to
have greater influence on both science and humankind than Albert Einstein (The Royal Society News,
2005). It would be interesting to find out whether that also holds true beyond the geography of the
UK.
2.3 Data
Our main datasets are Google digital books and scholarly articles. As shown in Table 1, the Google
corpus used in this study consists of over 36 million global digital books and 91 million articles
published since the 16th century.
(Table 1 insert here)
Table 1. Coverage of Google Books and Google Scholar
Google
corpus
Total
returned
hits
After 2000
20th century
19th
century
18th
century
17th
century
Before
1600
Books
36,540,000
14,600,000
14,700,000
5,680,000
808,000
319,000
433,000
Scholar
91,258,000
43,061,000
46,935,600
1,180,630
80,300
2
0
Note: This search was conducted through the library at London University in July 2015. For Google
Books, the command we used is “allinurl:books” at https://books.google.com. For Google Scholar
we set up the custom range of years in the advanced search. 3
To gain a complete picture of these physicists’ fame worldwide, we took a variety of languages into
consideration. We first tested our search terms in 57 different languages using the Google Books
search engine based on trials and errors.4 The volumes and shares of the written languages of these
books are displayed in Table 2. As shown, the dataset we analyzed covers more than 126 thousand
books mentioning Newton and 149 thousand books mentioning Einstein since the 20th century. This
3 It needs to be pointed out that depending on the search date and location, the returned hits vary. Take Google Books, for
example. We repeated the same search in July 2018, and the returned hits are not 36,540,000 as in July 2015 in the UK
but are 38,400,000 in China and 39,300,000 in South Korea. This pattern also holds for Google Scholar. As shown in
Table 1, the coverage of Google Scholar was 43,061,000 articles published over the period of 2000–2015 when the
search was conducted at London University in July 2015. In August 2018, its coverage of the same period extends to
46,840,000 and 48,720,000 articles when searched in China and South Korea, respectively. Our experiences echo
previous findings on the coverage scope, growth rate, and indexing quality of Google Books and Google Scholar
(Abdullah & Thelwall, 2014; Fagan, 2017; Halevi et al. 2017). For an extensive review of the pros and cons of Google
Books and Google Scholar as well as their utilities in research assessment, please refer to Google Scholar Digest at
http://googlescholardigest.blogspot.com/p/bibliography.html.
4 The variations of name order (e.g., “Isaac Newton” and “Newton, Isaac”) were taken into consideration.

4
finding seems opposite to that of the UK polls in 2005. Table 2 also reveals that, similar to other
publication datasets, the coverage of Google Books is also biased in language (Lin 2012; Liu, et al.
2018; Liu et al., 2015). Books written in Chinese, Russian, Korean, and other non-English languages
are highly underrepresented in the Google Books corpus. This caveat needs to be borne in mind
when using this search engine for analysis.
(Table 2 insert here)
3. Analysis
3.1. Global fame over time
Figure 1 depicts the quantity of digitalized books and academic items mentioning Newton’s and
Einstein’s full names. It needs to be pointed out that our returned hits consist of two scenarios: books
mentioning Newton or Einstein and books written by either of them. But given the number of
returned hits and the number of books even the most prolific scholar can write, it is reasonable to
believe that the majority of the retrieved books are those mentioning them. Section 4 discusses this in
more detail. Blue denotes Newton while red represents Einstein. The dotted lines mark the deceased
years of the two great scientists, respectively. Please note that any key word search is only effective
for accessible content. In our case, only Google Books with a preview available can be searched for
names in the text.
(Figure 1 insert here)
In Figure 1A, the number in the upper left (1e4=104) is the unit of the absolute number of books
indexed in the Google Books search. For Newton, the most glorious period of his historical influence
is between 1680 and 1880, when records mentioning Newton amount to more than 60% of all
records mentioning him. As demonstrated, the deaths of these great scientists do not dim their glory.
On the contrary, the popularity of both physicists has risen dramatically since the 1980s. One may
wonder what happened in the period of 1880 to 1980, when relatively fewer books talked about
Newton and Einstein. One speculation is that two dark clouds appeared in the world of physics at the
end of the 19th century: the Michelson–Morley experiment and the ultraviolet catastrophe in Planck’s
law (Kelvin, 1901). They overturned the framework of the old theory system, leading to the birth of
quantum and relativity theories (Hoover, 1977; Sanghera, 2011). This, when combined with the
breakout and aftermath of World Wars I and II, partially explains the silence of Newton’s voice from
1880 to 1980.
Figure 1B maps the scientists’ influences in academia via a Google Scholar search engine. The y
values, i.e. relative counts, were computed by dividing the number of Google Scholar indexed
articles mentioning the scientist in a given year by the total number of Google Scholar indexed
articles in that year. As shown, Newton leads in fame until 1918, then both scientists parallel for 30
years. The year of 1948 is the watershed. Since then, Einstein has gained more popularity than
Newton worldwide.
One key message conveyed by Figure 1 is that globally Einstein seems to enjoy greater fame than
Newton, which is contrary to the results of the 2005 UK polls mentioned previously. If the findings
of both the surveys and the computational analysis are valid, one possible explanation for this
discrepancy is that scientific fame varies by location. People are more recognized within their own
group (Egghe and Rousseau, 2004; Tang et al., 2015), be it by country or by shared language. To test

5
Table 2. Retrieved hits in Google Books by different languages
Scientist
Search word
Retrieved hits in
57 languages
2000–2015
(prop.)
1900–1999
(prop.)
Isaac Newton
Records in 49
languages with the
same name spelling
as in English
95.78%
98.18%
Исаак Ньютон
Russian
2.01%
1.28%
아이작 뉴턴
Korean
1.17%
0.12%
アイザック・ニュートン
Japanese
0.51%
0.31%
"艾萨克·牛顿"OR"艾薩
克·牛頓"
Chinese (simplified
or traditional)
0.21%
0.06%
ﻦﺗﻮﻴﻧ ﻕﺎﺤﺳﺇ
Arabic
0.18%
0.04%
ןוטוינ קזייא
Hebrew
0.15%
0.02%
ंंंंं ंंंंं
Bengali
0.01%
0.00%
ंंंंं ंंंंंं
Hindi
0.00%
0.00%
Isaac
Newton
(1643–1727)
Total retrieved hits of
Newton
125593
36434
Albert Einstein
Records in 49
languages with the
same name spelling
as in English
96.25%
96.78%
Альберт Эйнштейн
Russian
2.35%
2.08%
ןייטשנייא טרבלא
Hebrew
0.42%
0.62%
アルバート・アインシュ
タイン
Japanese
0.35%
0.22%
알버트 아인슈타인
Korean
0.26%
0.08%
"阿尔伯特・爱因斯坦
"OR"阿爾伯特・愛因斯坦"
Chinese (simplified
or traditional)
0.21%
0.17%
ﻦﻳﺎﺘﺸﻨﻳﺍ ﺕﺮﺒﻟﺍ
Arabic
0.05%
0.06%
ंंंंंंं
ंंंंंंंं
Hindi
0.01%
0.00%
ंंंंंंं
ंंंंंंंंं
Bengali
0.00%
0.00%
Albert
Einstein
(1879–1955)
Total retrieved hits of
Einstein
149169
38154
Data source: Google Books search. Data accessed in January 2016.
Note: In the other 49 languages such as German, French, Italian, Spanish, and Portuguese, both name
spellings are exactly the same as in English.
this speculation, we utilized the language option of Google Books Ngram Viewer and examined the
physicists’ historical influence by various languages.5
5 Since 2010 the Google Books Ngram Viewer has been available at http://books.google.com/ngrams. Its source, i.e.
Google Books, was generated through Partner Program and the Library Project (please refer to
https://www.google.com/intl/en/googlebooks/about/). These millions of books published between 1500 and 2008 were
digitally scanned and the corpus was winnowed. For more details please see Michel et al. (2011) and Twenge et al.

6
Figure 1. Global fame of Newton and Einstein
Data source: Panel A is Google Books search. The time coverage of the main
graph is from 1680s to 2010s with an inset zooming in on the period of 1880s–
1980s; Panel B is Google Scholar search. The time coverage is from 1900 to 2012.
Accessed in January 2016.
Forty-nine languages with the same name spelling as in English are included.
3.2. Global fame moderated by language
Figure 2 was generated with Google Books Ngram Viewer with a default smoothing of three.6 The
four panels illustrate the scientific fame of Newton and Einstein in books written in British and
American English together, British English alone, American English alone, and German, respectively.
Both scientists seem to be preferred in their own groups. In the British English case, Newton has
consistently enjoyed more popularity than Einstein throughout the last 100 years (Panel B). This
finding is consistent with 2005 UK polls results. In sharp contrast, Einstein, a German-origin
physicist who later immigrated to the US, has been mentioned far more than Newton since the mid or
late 20th century in British and American English together, American English alone, and German
books (Figures 2A, 2C, and 2D). This comparison of citing book language uncovers a high relevance
of own-group preference, or approximately location of fame.
(Figure 2 insert here)
(2012). The searchable corpora with the Google Books Ngram Viewer were developed by the Google Team and can be
downloaded at http://storage.googleapis.com/books/ngrams/books/datasetsv2.html. Please note that Google Books
Ngram Viewer cannot be used to determine how often a search term occurs in a book.
6 The default smoothing parameter of Google Books Ngram Viewer is set at 3. Taking the year of 2000 as an example, a
smoothing of 3 means that the data shown for 2000 are a mathematical average of original data for 2000 plus three values
before and three after 2000 (i.e., a moving average of seven-year data, including 1997, 1998, 1999, 2000, 2001, 2002,
and 2003). This setting avoids spikes and makes trends more apparent (Michel et al., 2011; Greenfield, 2013). More
information on graphs generated by Google Books Ngram Viewer is available at https://books.google.com/ngrams/info.

7
Figure 2. Scientific fame by language
Data source: Google Books Ngram Viewer. Accessed in January 2016.
Panel A: Only books in English (both British English and American English) are considered.
Panel B: Only books in British English are considered.
Panel C: Only books in American English are considered.
Panel D: Only books in German are considered.
The time coverage is from 1800 to 2008 for Panel A and from 1900 to 2008 for Panels B~D.
3.3 Achievements they are famous for
Scientists are commemorated for various reasons. Hawking and Israel (1989) note that the greatest
contributions made by Newton in the history of science are the law of universal gravitation and the
three laws of motion. In addition to physics, Newton also contributed to the fields of mathematics,
astronomy, and the philosophy of nature (Newton, Cohen, & Schofield, 1959; Newton, 1999). In
comparison, the most notable achievements of Einstein are quantum mechanics and general relativity
(Whittaker, 1943; Torre, et al., 2000). Then, a question arises: what are they famous for as evidenced
by the Google corpus?
As demonstrated in Figure 3A, the contributions of Newton surpass the Laws of Motion and the Law
of Gravity. This is particularly true prior to mid-1800. As shown the name of Isaac Newton is
mentioned in Google Books far more than his most accredited physics theories are.

8
(Figure 3 insert here)
Figure 3. The major scientific contributions of Newton and Einstein
Data source: Google Books Ngram Viewer. Accessed in December 2017.
The time coverage is from 1700 to 2008 for Panel A and from 1900 to 2008 for Panel B.
One may notice the close intertwining or convergence of lines representing scientific achievements
and the scientists’ names in later periods. But the approximately same frequency of mentioning the
achievements and the scientists’ names does not mean that the mentions appear simultaneously
within the same corpus books. So, we next conducted co-occurrence analysis to examine to what
extent these scientists’ most accredited achievements contribute to their enduring reputations (Table
3).
(Table 3 insert here)
As depicted in Table 3, books mentioning Isaac Newton often deal with the law of universal gravity
(15%) and calculus (8%). The influence of Albert Einstein is highly correlated to the theory of
relativity (about 28%) and quantum theory (almost 17%). This, on one hand, reveals what
achievements contribute to the glory of Newton and Einstein; on the other hand, the residue (i.e., the
uncorrelated proportion) suggests other elements (such as other achievements or even anecdotes)
exist.7
7 Our manual examination of the remaining books mentioning Newton but not gravity, gravitation, calculus, or motion
suggests that many are related to natural philosophy, alchemy, and theology. In addition to mentions of relativity,
quantum, photoelectric and mass-energy, books mentioning Einstein also discuss his ethical and philosophical views on
cosmic religion, his educational philosophy, and even his private life.

9
Table 3. Co-occurrence analysis of contributions and physicists
1900–2015
Key words
Records in Books
% of Total
Isaac Newton
156,100
/
Newton and gravity
(including law of gravity, universal gravity, universal
gravitation, law of gravitation)
23,700
15.2%
Newton and calculus
12,880
7.9%
Newton and laws of motion
10,440
6.7%
Albert Einstein
178,200
/
Einstein and relativity
(including general relativity and special relativity)
50,130
28.1%
Einstein and quantum
30,020
16.9%
Einstein and photoelectric
4,200
2.4%
Einstein and mass-energy
2,420
1.4%
Data source: Google Books search. Data accessed in January 2016.
3.4 Other great minds in physics
We extended our analysis to other great physicists in history. We first compiled a list of 234 of the
most influential scientists based on the following four books:
The 100 Most Influential Scientists of All Time (Rodgers 2009)
The 100 Most Influential Scientists (Simmons 2009)
100 Scientists Who Changed the World (Tiner 2002)
The Most Influential Scientists in Human Civilization (Luo 2012)
After removing scientists born before the year 1 BC, such as Aristotle and Plato, we next searched
each name in Wikipedia and allocated the scientists to different research fields. We then retrieved
their name appearance frequency in all Google Books published after 2000. The top 20 physicists
listed in Table 4 show that in the field of physics, Albert Einstein, Max Planck, Isaac Newton, Blaise
Pascal, and Galileo Galilei are the top five most influential and best-known physicists in the 21st
century.
(Table 4 insert here)
Table 4 lends further support that great scientists are gone but not forgotten.8 As shown, the early
scientists such as Newton, Pascal, and Galilei are still on the public’s lips in contemporary society.
Undoubtedly, the achievements of these scientific pioneers have become the cornerstones of human
scientific knowledge. But it is also true that besides scientists’ contributions to scientific
advancement, their scientific spirit, anecdotes, and other special attributes may remain as well.
Students and the public learn about the great scientists via scholarly publications, textbooks, and
popular science books. When a scientist becomes a public figure, his life experiences and
8 The scientific fame of these 20 physicists persists after they pass away.

10
achievements are often adapted into various biographies and even enlightening stories. In stories
such as “Newton and His Apple,” the scientific spirit has been used to encourage and to inspire each
succeeding generation for hundreds, if not thousands, of years.
Table 4. Top 20 physicists ranked by scientific fame
Rank
Name of Physicist
Nationality
of Origin
Year of
Birth
Rec. of
being
mentioned
in Books
Rank
Name of Physicist
Nationality
of Origin
Year
of
Birth
Rec. of
being
mentioned
in Books
1
Albert Einstein
German
1879
145,000
11
Michael Faraday
UK
1791
49,500
2
Max Planck
German
1858
129,000
12
James Watt
UK
1736
47,900
3
Isaac Newton
UK
1643
121,000
13
Clerk Maxwell
UK
1831
43,100
4
Blaise Pascal
France
1623
118,000
14
Enrico Fermi
Italy
1901
41,400
5
Galileo Galilei
Italy
1564
96,100
15
William Thomson
UK
1824
40,700
6
Stephen Hawking
UK
1942
82,800
16
Robert Hooke
UK
1635
40,200
7
Niels Bohr
Denmark
1885
80,100
17
Max Born
German
1882
38,000
8
Joseph Henry
USA
1797
79,100
18
Wolfgang Pauli
Austria
1900
37,200
9
Richard Feynman
German
1918
65,300
19
Robert Oppenheimer
USA
1904
36,400
10
Werner Heisenberg
German
1901
55,400
20
Thomas Young
UK
1773
36,400
Data source: Google Books search. Data accessed in January 2016.
Time coverage: 2000–2015
This also holds for Stephen Hawking. In the list of top 20 physicists in the 21st century shown in
Table 4, Hawking ranks sixth and far ahead of some classic scientists including James Watt, Michael
Faraday, and so on. In addition to Hawking’s great scientific achievements, his passion in promoting
science given his special physical condition has been seen as inspiring to the public and especially
younger generations.
4. Conclusion and discussion
4.1 Summary
Big data is an important theme in natural and social sciences today. Adopting and expanding the
approaches developed by Bohannon (2011a, 2011b), Michel et al. (2011), and Lin et al. (2012), this
work exemplifies how to track the historical achievements and influence of great scientific minds.
Utilizing the language option of Google Books Ngram Viewer, we find evidence in support of own-
group preference, which explains the discrepancy of the UK survey and computational analysis
results based on the Google corpus. The co-occurrence analysis demonstrates the main contributions
that scientists are known for by later generations
The wheel of history is always moving forward, with new developments emerging and old theories
gradually fading away. This is an inexorable law of history, with each scientist’s face and influence
having a certain vitality cycle. Yet our study suggests the great minds live long intellectually. A
scientist can only live several decades in physical form, but his contributions can live on in human
history for hundreds or even thousands of years. Today when we talk about pioneers of science
before the Christian era such as Aristotle, Euclid, and Archimedes, we find that these names have
been written into history, respected and eulogized repeatedly by generations of newcomers. This may
be the highest level of the meaning of life that scientists are pursuing.

11
4.2 Future research
It is an accepted practice to use citations to measure the scientific impact of researchers or their
research output. Yet for an academic paper, the citation life cycle is usually about 3 to 5 years, and
the count of citations often declines drastically afterwards (Glänzel and Schoepflin 1995; Tang 2013;
Rogers 2010). For a long time bibiometricians have used or abused this impact factor while also
acknowledging the difficulties in quantifying long-term scientific impact (Wang, Song and Barabási,
2013; Waltman 2016). With the development of Google digitalized texts and computational
analytical methods, we can track names being mentioned and assess the evolution of scientists’ fame
on a global scale over centuries. Given the variety of book types indexed in Google, this opens
another possible research venue of altmetrics: measuring the influence of scientists beyond academia
(Costas et al., 2015).
In his critical overview of altmetrics, Bornmann (2014) pointed out that no accepted framework
existed to measure societal impact. Different from the existing altmetrics, which assess the impact of
scientific work mostly on social media, we argue that when applied appropriately, Google Books can
serve as a promising source of data for altmetrics, and the appearance frequency of scientists’ names
in books can serve as alternative metrics, which are still in great flux, to capture scientists’ intel-
lectual contributions to a broader public (Bornmann, 2014; Fenner, 2013; Piwowar, 2013). In the
future it would be interesting to differentiate the fame of a scientist within academia and among the
public with the categories provided in Google Books search (textbook, fiction, scholarly book, etc.).
This paper focuses on physics; research looking into scientific fame of figures in other disciplines or
even across disciplines is also worthy of further examination.
4.3 Limitations
In spite of their potential value for evaluation research, we would like to raise several notes of
caution regarding the data and the method.
There are two sources of false positives. Searching individuals’ names in books or scholarly
publications can yield two types of results: works written by them and their achievements cited or
mentioned by others. Given that Google Books and Ngram Viewer cannot differentiate the specific
locations of search terms within books, false positives are inevitable. The good news is that even for
an extremely prolific scientist, the actual number of articles one can write and publish is only in the
hundreds. As an illustration of authorship versus citation, we searched Alexandre Dumas (1802–
1870) who, though not a scientist, was a prolific French writer who dedicated his life to writing and
published 357 books. We searched his full name in Google Books on December 20, 2017, and
retrieved 23,400 hits. This suggests that the number of times mentioned can serve as a proxy
indicator of fame of prominent individuals and is not a reflection of their authorship.
The other source of false positives is name disambiguation. For instance, searches for just “Newton”
could result in mentions of other people or even the International System of Units (SI) unit of force.
The common name problem, translation, and transliteration can also lead to an overestimation of
scientists’ fame. On the other hand, name changes over time associated with marriage and other
reasons can produce false negatives, or an underestimation of reputation. In this sense the
measurement of scientific fame works best for great scientists or scientists with unique names. For
more details, please refer to MacRoberts and MacRoberts (1989) and Tang and Walsh (2010).
We also need to be aware of the issues of eponymy and obliteration, which have been discussed
rather comprehensively by McCain (2011). When a scientist’s achievements are so important as to
become common knowledge, it is possible that only the achievements themselves rather than the
scientist’s full name are referenced. For instance, when people talk about the law of gravity, they
may not mention Newton’s name at all. In this case, using full name as a proxy causes an
underestimation of fame. As McCain noted, “it is often observed that very well-established concepts,

12
experimental methods, empirical findings, and models are mentioned in the literature without being
linked to their originators” (p. 1413). This in our case will incur false negatives of recording
scientists’ fame. However, we argue that a scientist’s full name never being mentioned once
throughout a whole book is rare and thus has minimal impact on our findings.
It should be noted that for any computational analysis the amount of retrieved data should be
sufficiently large. Scientific fame is no exception. Despite its potential, researchers should be aware
of the caveats of the Google corpus, such as language bias against non-English languages, the quality
heterogeneity of indexed books, and whether it is appropriate only for highly influential scientists
rather than general scientists, when conducting fame analysis. It needs to be pointed out that though
Google Books and Ngram Viewer make it possible for us to analyze already famous scientists’ fame,
it would not be possible to identify those who are not yet famous scientists.9
Another caveat of this novel measurement is that it may not separate mentions of a famous figure for
infamy rather than mentions of the figure for positive glory. Fortunately, in our case of two well-
acknowledged great minds in science, the majority of the books on Newton and Einstein are based on
a positive or at least a neutral evaluation. This saves us from the concern about evaluating
controversial figures.
Acknowledgements
The research is supported by the National Social Science Foundation of China (#14CXW011), Korea
Foundation for Advanced Studies (International Scholar Exchange Fellowship 2017-2018), and the
Humanities and Social Sciences Project of the Ministry of Education of China (#WKH3056005).We
would like to express our sincere gratitude to two anonymous reviewers and Professor Ludo
Waltman for their insightful suggestions which substantially improved this manuscript. We thank
Professor Fei Xu for the inspiring discussion, Jiafei Shen and Feifei Han for their help in handling
the figures. Author contributions: G.W., C.L., and L.T. designed research; G.W. analyzed data; and
L.T and G.H. wrote the paper. The authors declare no conflict of interest.
9 We would like to thank one anonymous reviewer for directing us to this point.

13
References
Abdullah, A., & Thelwall, M. (2014). Can the Impact of non-Western Academic Books be Measured?
An investigation of Google Books and Google Scholar for Malaysia. Journal of the Association
for Information Science and Technology, 65(12), 2498–2508.
Acerbi, A., Lampos, V., Garnett, P. & Bentley, R.A. (2013).The Expression of Emotions in 20th
Century Books. PLOS ONE 8(3): e59030
Astin, A.V. (1957). Scientists and public responsibility. Physics Today 10(11), 23-27.
Baker, E.P. (1984). Einstein versus Newton: the gravity of the situation. Ohio: Online Computer
Library Center.
Bohannon, J. (2011a). The science hall of fame. Science, 331(6014), 143-143.
Bohannon, J. (2011b), Google Books, Wikipedia, and the Future of Culturomics. Science, 331(6014),
135-135.
Bornmann, L. (2014).Do altmetrics point to the broader impact of research?An overview of benefits
and disadvantages of altmetrics. Journal of Informetrics, 8(4), 895-903.
Bronk, D. W. (1976). Alfred Newton Richards (1876-1966). Perspectives in Biology & Medicine,
19(3), 413.
Costas, R., Zahedi, Z., & Wouters, P. (2015). Do “altmetrics” correlate with citations? Extensive
comparison of altmetric indicators with citations from a multidisciplinary perspective.
Journal of the Association for Information Science and Technology 66(10), 2003-2019.
Egghe, L. & R. Rousseau (2004). How to measure own-group preference? A novel approach to a
sociometric problem. Scientometrics 59(2), 233-252.
Fagan, J.C. (2017). An evidence-based review of academic web search engines, 2014-2016:
Implications for librarians’ practice and research agenda. Information Technology and Libraries,
36, 2
Fenner, M. (2013). Letter from the guest content editor: Altmetrics have come of age. Information
Standards Quarterly, 25(2), 3-3.
Feist, G. J. (2016). Intrinsic and extrinsic science: A dialectic of scientific fame. Perspectives on
Psychological Science, 11(6), 893-898.
Glänzel, W, & Schoepflin, U., (1995). A bibliometric study on aging and reception processes of
scientific literature, Journal of Information Science, 21, 37–53.
Graneau, P., & Graneau, N. (1993) Newton versus Einstein: How matter interacts with matter.
Sheffield: Carlton Press.
Greenfield, P.M. (2013).The Changing Psychology of Culture From 1800 Through 2000.
Psychological Science 24(9), 1722-1731.
Gribbin, J. (1987). Newton vs. Einstein. New Scientist 116(1586):45-45.
Hoover, E.R.(1977). Cradle of Greatness: National and World Achievements of Ohio’s Western
Reserve. Ohio: Cleveland.
Hawking, S. W., & Israel, W. (Eds.). (1989). Three hundred years of gravitation. Cambridge
University Press.
Halevi, G., H. Moed and J. Bar-Ilan (2017). Suitability of Google Scholar as a source of scientific
information and as a source of data for scientific evaluation—Review of the
Literature.Journal of Informetrics 11(3), 823-834.
Laubichler, M.D., Maienschein, J., & Renn, J. (2013). Computational Perspectives in the History of
Science: To the Memory of Peter Damerow. ISIS 104(1): 119-130.
Liu, W., et al., (2015).China's global growth in social science research. Journal of Informetrics, 9(3),
555-569.
Liu, W., et al., (2018). Author address information omission in Web of Science —An explorative
study. Journal of Informetics, forthcoming.

14
Luo, J.W. (2012). The Most Influential 100 Scientists in Human Civilization. DAyi Publishing,
Xiamen (in Chinese)
Macroberts, M. H., & Macroberts, B. R. (1989). Problems of citation analysis: A critical review.
Journal of the American Society for Information Science, 40, 342–349.
McCain, K. W. ( 2011) Eponymy and Obliteration by Incorporation: The Case of the “Nash
Equilibrium”. Journal of the American Society for Information Science and Technology, 62
(7),1412–1424.
Menard, H.W. (1971).Science: Growth and Change. Harvard University Press, Cambridge, MA.
Merton, R. K. (1970). Behavior patterns of scientists. American Scientist, 3(2), 213-220.
Michel, J. B., Shen, Y. K., Aiden, A. P., Veres, A., Gray, M. K., Pickett, J. P., .& Pinker, S. (2011).
Quantitative analysis of culture using millions of digitized books. Science 331(6014), 176-
182.
Newton, I. (1999). The Principia: mathematical principles of natural philosophy. University of
California Press.
Newton, I., Cohen, I. B., & Schofield, R. E. (1959). Isaac Newton's papers and letters on natural
philosophy.
Pettit, M. (2016).Historical time in the age of big data: cultural psychology, historical change, and
the google books ngram viewer. History of Psychology 19(2), 141-153.
Piwowar, H. (2013). Altmetrics: Value all research products. Nature, 493(7431), 159-159.
Rogers, J.D. (2010).Citation analysis of nanotechnology at the field level: implications of R&D
evaluation. Research Evaluation 19(4), 281-290.
Roth, S. (2013).The Fairly Good Economy: Testing the Economization of Society Hypothesis
Against a Google Ngram View of Trends in Functional Differentiation (1800-2000). Journal
of Applied Business Research 29(5),1495-1500.
Sanghera, P. (2011). Quantum Physics for Scientists and Technologists: Fundamental Principles and
Applications for Biologists, Chemists, Computer Scientists, and Nanotechnologists. Wiley-
Blackwell.
Simmons, J.G. (2009). The 100 Most Influential Scientists Paperback. New York: Fall River Press.
Sternfeld, J. (2011). Archival theory and digital historiography: Selection, search, and metadata as
archival processes for assessing historical contextualization. The American Archivist 74(2),
544-575.
Stöver, Stöver, D. J. H., Trapp, J., Stoever, D. H., & Linné, C. V. (1794). The Life of Sir Charles
Linnæus. London, E. Hobson. Library of Congress. ISBN 978-0-19-850122-0.
Tang, L. (2013). Does "birds of a feather flock together" matter-Evidence from a longitudinal study
on US-China scientific collaboration. Journal of Informetrics 7(2), 330-344.
Tang, L., & Walsh, J.P. (2010). Bibliometric Fingerprints: Name Disambiguation Based on
Approximate Structure Equivalence of Cognitive Maps. Scientometrics 84 (3), 763-794.
Tang, L., Shapira, P. & J. Youtie (2015). Is there a clubbing effect underlying Chinese research
citation increases? Journal of the Association for Information Science and Technology 66(9):
1923–1932.
Tiner, J.H. (2002).100 Scientists Who Changed the World. New York: Gareth Stevens Publishing.
Torre, D.L., Daleo, A., &I. García-Mata (2000). The Photon-Box Bohr-Einstein Debate
Demythologized. European Journal of Physics 21 (3): 253-260.
Twenge, J., Campbell, K. W. & Gentile, B . (2012). Increases in Individualistic Words and Phrases
in American Books: I960 - 2008. PLoS ONE 7 (7) :e40181-e40181
Waltman, L. (2016). A review of the literature on citation impact indicators. Journal of Informetrics
10(2), 365-391.
Wang, D., Song, C., & Barabási A.L., (2013). Quantifying long-term scientific impact.
Science 342 (6154): 127-132.

15
Whittaker, E.T. (1943). Aristotle, Newton, Einstein, Science 98(2542), 250-254.Youtie, J., Rogers, J.,
Heinze, T., Shapira, P., & Tang, L. (2013). Career-based influences on Scientific
Recognition in the United States and Europe: Longitudinal Evidence from Curriculum Vitae
Data. Research Policy, 42(8), 1341–1355.
Zeng, R.& Greenfield, P.M. (2015). Cultural evolution over the last 40 years in China: Using the
Google Ngram Viewer to study implications of social and political change for cultural values.
International Journal of Psychology 50 (1), 47-55.