Reputations ebb and flow, leaving little trace of the course they followed in the historical record. Plato (427–347 B.C.) [25] and Aristotle (384–322 B.C.) [8] reign today as giants of Western civilization, but it was a very, very close call for both of them. With the descent of the Dark Ages after the fall of the Roman Empire, the writings of the ancient Greek philosophers were almost lost forever.

People's reputations often change posthumously. Paul Revere (1735–1818) [627] was a silversmith rescued from obscurity by Henry Wadsworth Longfellow's (1807–1882) [349] poem “The Midnight Ride of Paul Revere.” Today, Harriet Beecher Stowe (1811–1896) [449] is far better known than her brother Henry Ward Beecher (1813–1887) [2905], but this wasn't the case during their lifetimes.

There is more to our computational study of history than significance rankings. Here we add another statistical arrow to our quiver. Google Ngrams is an amazing resource for monitoring changes in the cultural Zeitgeist, a dataset that lets us reconstruct how famous someone was 50, 100, or even 200 years ago. In this chapter, we introduce Google Ngrams and present the graphical representations we will use to make proper sense of this data. In particular, we will use it to study the phenomenon of posthumous changes in fame.

Google Ngrams

Scholarly works often include many pages of footnotes, with citations proving that the author has read hundreds of books. Ours doesn't, which might suggest Philistine standards of scholarship.

Sports captures something important about human culture and society. Literary critic Wilfrid Sheed (1930–2011) [294893] observed that “sports communicate a code, a language of the emotions, and a tourist who skips the stadiums will not recoup his losses at Lincoln Center or Grant's Tomb.”

Even if you don't give a hoot about sports, skim though this chapter. Think of it as crib notes for a test on cultural literacy. Who are the most significant male and female athletes of all time? The leading figures in boxing, football, hockey, tennis, and golf? What makes James Naismith (1861–1939) [847] a more important basketball figure than Michael Jordan (1963–) [1115]?

Sports is a domain that generally interests men much more than women. Yet we will identify one professional sport where basic gender equality holds: its ten most significant athletes are equally split between men and women.

Individual Sports

We start by looking at professional sports where athletes compete as individuals: boxing, tennis, golf, and horse racing. We defer analysis of historically amateur Olympic sports like track and field to Section 12.3.

BOXING

Boxers appear to have been the earliest class of professional athletes to achieve widespread renown. The earliest bare-knuckle champion of England was James Figg (1695–1734) [31351], who won the title in 1719. Boxing was a particularly rough business before the 1860s, when the sport became standardized by the Marquess of Queensberry rules (established by John Douglas (1844–1900) [9223]). The shift from bare-knuckle to gloved boxing followed shortly thereafter, with John L. Sullivan (1858–1918) [4542] being the major transitional figure.

Helping your children with their homework is a part of every parent's job description. It can be frustrating to watch your kid struggle with something as basic as elementary subtraction, but incredibly rewarding when they finally master borrowing ten under your tutelage.

Homework gets harder as students progress through grade levels and, like all parents, I (Steve) realized that the time would come when I would no longer be able to help my daughter Bonnie with her homework. But I didn't expect that moment would first arise in fifth grade U.S. history!

Bonnie's assignment was to match historical figures with descriptions of who they were and why they were important. Several names were well known to her from prior experience, such as Christopher Columbus (1451–1506) [20], George Washington (1732–1799) [6], and Abraham Lincoln (1809–1865) [5]. Others were new, but clearly worth learning about as she began her study of American history – people such as:

1. • Henry Ford (1863–1947) [148]: automobile industry pioneer.

2. • Robert Fulton (1765–1815) [954]: steamboat inventor.

3. • William Lloyd Garrison (1805–1879) [735]: abolitionist who fought against slavery.

4. • Alexander Hamilton (1755–1804) [45]: secretary of the Treasury who put the United States on a firm financial footing.

5. • Robert E. Lee (1807–1870) [76]: commander of the Confederate Army during the Civil War.

But Bonnie's homework also contained the names of several people who, frankly, I had never heard of before.

Our rankings of historical significance will be used to rationally assess the reputation of figures in the historical canon. But how can algorithms do this in a fair and sensible way? We start by reviewing how people approach ranking problems, to better understand the strengths and limitations of our computational methods.

Traditional Ranking Methodologies

The rankings most prominent in popular culture are produced in several ways. Perhaps most popular is the expert poll, used to create the Associated Press Top 25 College Football Rankings. Every week a set of experts (here coaches and sportswriters) independently rank the top teams. These ratings are combined using a point system, and the cream skimmed off after sorting teams by points yields the Top 25. Expert polls prove effective in clearly defined domains followed by many knowledgeable individuals. They tend to be fairly conservative, however, and are often based on second-hand judgments. Do we really believe that active college football coaches have enough time, or even a good enough cable television plan, to watch all their peers play?

Still, polls of professional historians might be used to rank historical figures. Indeed, historian polls ranking the greatness of U.S. presidents are conducted regularly as a sort of academic parlor game; an early example was the 1948 poll by Arthur M. Schlesinger (1888–1965) [27072] of Harvard University. These polls reflect interesting changes in historical reputation over time.

The battle between good and evil has been fought since the beginning of time. Which is the surer path to historical recognition? There seems to be conflicting evidence wherever you look:

1. • For every Jesus (7 B.C.–A.D. 30) [1], there is an Adolf Hitler (1889–1945) [7].

2. • For every charitable missionary like Mother Teresa (1910–1997) [820] lurks a gangster like Al Capone (1899–1947) [646].

3. • For every social worker like Jane Addams (1860–1935) [1256] exists an assassin like Lee Harvey Oswald (1939–1963) [1435].

4. • For every civil rights leader like Martin Luther King (1929–1968) [221], we find a serial killer like Jack the Ripper [166].

In this chapter, we will consider a range of figures from both ends of the morality scale, to open a discussion on which is the more enduring path to glory. Our conclusion is that true virtue is generally rewarded, but spectacular acts of infamy endure longer than similar acts of heroism.

The Killers

Hammurabi (1796–1750 B.C.) [899], the lawmaker, has been followed by three millennia of law breakers. But several criminals richly deserving of their punishment endure as cultural figures because of the imagination and magnitude of their crimes. In this section we will identify the most prominent outlaws, assassins, and other killers, to better understand the source of this fascination.

The previous chapter presented the popular history of American political leadership. Here we apply our analytical tools to study the reputations of the rest of the world's leaders. We are particularly interested in understanding the factors that shape our perceptions of other places, such as size, economics, and culture.

Knowing the who's who of a country is helpful to understand what's what. In the same way that speaking a trivial amount of the local language (“excuse me”) can pay big dividends, we have found that knowing the name of the current or iconic political leader opens doors to greater understanding. Asking immigrants or locals “What do you think of …?” is an excellent way to start a revealing conversation. We provide tables identifying the dominant political figure from essentially every country on earth. These can be seen as crib notes in cultural literacy; a chance to refresh our memories about the people who built or led nations.

Analyzing Political Leadership

In the course of our study, we have created a dataset of more than 3,000 modern national leaders from more than 150 different countries. But any meaningful attempt to identify the “most significant” leader of a nation faces several issues of definition and bias. We discuss some of these considerations here.

CULTURAL AND REGIONAL BIAS

Our analysis relies on datasets that inherently display linguistic and cultural biases. Both our Wikipedia analysis and book Ngrams data are based exclusively on English-language sources, and hence do not equally reflect the contributions of other languages and cultures.

Websites and Databases

We hope that some readers will be inspired by this book to perform further investigations into historical significance. We encourage you to visit our website at http://www.whoisbigger.com, where you can access our latest analysis for more than 800,000 different people. These rankings will be updated to reflect more recent editions of Wikipedia, including those of languages beyond English.

This website is still under active development, so you will have to visit to see exactly what information we present on each individual. But at the time of this writing, we include rosters of people with comparable significance, celebrity, and gravitas, to help you put their achievements into perspective.

We also provide an interface to rank individuals within hundreds of thousands of Wikipedia categories, so you can see how they stack up in comparison with their peers.

We also encourage the reader to check out the Google books Ngram viewer, which Google makes available at http://books. google.com/ngrams. Although we have developed our own Ngram visualizations, which we make available at http://www.whois bigger.com, the Google site has far more comprehensive data. This includes more general vocabularies, annotation by parts of speech, and even Ngrams in multiple languages. It is a fabulous resource to explore.

Birth/Death Calendars

In Section 7.2 of Chapter 7, we showed that life outcomes can be affected (at least to a minor extent) by the specifics of one's calendar birth date. It is natural to be curious about the historical figures which share your birth date: perhaps this kinship may reflect something of our own prospects for the future.

Most historical figures are accorded a one-way ride down the road from glory. Long-dead figures rarely regain prominence, except perhaps in response to external events, such as the release of a movie about their life. But the most dramatic swings in historical reputation we have encountered occur in the arts, reflecting long-term fluctuations in taste. Authors and their books can rise in response to fresh critical attention, and obscure painters may suddenly find their work back in style.

In this chapter, we will identify the most significant figures in literature and the fine arts. Along the way we will address certain larger-scale questions: In which media do contemporary artists have enough stature to compete with the classical masters? How well do best-seller lists and literary awards identify the significant voices of their times? Have modern architects usurped the role traditionally occupied by sculptors?

Literature

Ranking the world's greatest literary figures is a parlor game of comparable popularity to the ranking of presidents. It provides free rein to express the biases inherent in everyone's worldview. Which is better: classical or contemporary, domestic or international, or prose vs. verses?

The Literary 100 [Burt, 2000] is a representative ranking, which tries to identify “the most influential novelists, playwrights, and poets of all time.” Burt provides an alternate perspective to evaluate our “Literary 50” rankings, presented in Figure 13.1.

Quantitative analysts (“quants”) are finding new homes in the social and cultural domains. Finance? Hedge funds employing quantitative trading strategies now rule Wall Street, to the extent that more than 80 percent of all buy-sell decisions are made by algorithms, not people. Sports? The best selling book Moneyball detailed how successful major league teams now hire general managers who value statistical analysis above gut instinct – and get to be played by Brad Pitt in the movie version of the story. Politics? Prediction markets and the meta-analysis of polling data work so reliably that it hardly seems necessary to hold the actual election.

In this book, we bring quantitative analysis to bear on ranking and comparing historical reputations. Who's bigger: Washington or Lincoln? Hitler or Napoleon? Picasso or Michelangelo? Charles Dickens or Jane Austen? Did you realize that:

1. • Although Paul Revere (1735–1818) [627] and Betsy Ross (1752–1836) [2430] are well known to all American schoolchildren, they fell into complete obscurity for several generations after their contributions to the American Revolution. Their rediscoveries, completely independent of their actual achievements, tell us much about the capricious forces of history.

2. • Women remain significantly underrepresented in the historical record compared to men. We can prove that women have long required substantially greater achievement levels (analogous to about 4 IQ points in the mean) than men to get equally noted for posterity.

Here we provide greater technical detail concerning our historical significance ranking methods. We will discuss (1) exactly what data we used, (2) how the data was normalized, and (3) the statistical methodology we used to develop our rankings. Interested readers should begin with the high-level overview of our methods that we presented in Chapter 2.

Feature Set

We used the October 11, 2010 Wikipedia release (http://dumps.wikimedia.org/enwiki/20101011/) as the basis for our analysis. We extracted six input variables from this distribution and other available datasets:

1. • Page Rank (NPR and PPR) – Wikipedia is a hyperlinked document, defining a network with vertices corresponding to articles and directed edges (x, y), meaning that article x references article y. The well-established PageRank measure of network centrality [Brin and Page, 1998] computes a significance score based on the number and strength of the in-links to each node. We compute two forms of PageRank, based on two different graphs derived from Wikipedia. The first contains all Wikipedia pages, while the second consists only of the pages corresponding to people. Page annotations were determined using Freebase (http://www.freebase.com/), a collaborative knowledge database. We employed the Cloud9 map-reduce library (https://github.com/lintool/Cloud9) to compute Page-Rank.

2. • Page Hits (PH) – Web logging data reveal how often eachWikipedia page is viewed. More famous/significant entities should have their pages read more frequently. We analyzed six months of log data, collected immediately prior to the date of our Wikipedia release. In order to reduce the large monthly variance in readership owing to news events, we used the median monthly frequency as our measure of page hits.

For thousands of years, man has watched stories told through flickering images, be they in front of a campfire or displayed on a television. Performance modes and styles have changed, but the human need to entertain and be entertained has remained a constant of our species.

The performing arts are both the oldest forms of human expression and the greatest contemporary source of new cultural memes. Primitive man and the hippest club-goers both loved music and dance, although it is hard to tell whether they would appreciate each other's art.

In this chapter, we will review the most significant people in the performing arts. We start from the era before recorded media and continue through the present day. We will analyze trends in classical and popular music, for clues as to which of these cultural streams is likely to dominate historically. We do the same for broadcast and visual media, studying the history of film, radio, and television as a guide to the future of communications.

Before Recording Technology

Sunday preachers can be thought of as early America's most prominent class of performing artist. The rigors of the six-day workweek and the strictures of the Christian Sabbath left little time for secular entertainment, but spellbinding ministers drew large crowds for Sunday worship. In his autobiography [Franklin, 1818], Benjamin Franklin (1706–1790) [35] estimated that more than 30,000 people attended a sermon by George Whitefield (1714–1770) [1258] in Philadelphia.

I (Steve) recently turned the big 5-0, a lifetime long enough to measure up against historical time scales in a nontrivial way. The grand old game of baseball's American League was founded in 1901, exactly 60 years and two days prior to my own founding. We have shared almost half of its history together. I recall the United States Bicentennial in 1976, amazed that our country was really 200 years old. Now I have personally witnessed more than 20 percent of American history since the Declaration of Independence. The Jewish calendar marks this year as 5774, meaning that my life will soon cover 1 percent of the span since the world's biblical creation. Young readers may smirk, but if you are lucky someday this will happen to you.

Time is the basic organization of history, measured across many scales: the calendar year, the human life span, and the historical age. But the flow of time seems deceptively slow moving, and most of it occurred off of our watch. We need calendars, timelines, and other tools to chart the course of history, and help us understand it better. In this chapter, we will analyze historical figures across different scales of time: ranging from the days of the year to vast historical epochs.

Life-Span Analysis

All men (and women) are created equal, in that they have but one life dedicated to the pursuit of happiness and achievement. But life spans are not equal, differing greatly in length and the period in which they are lived. There are several revealing ways to look at history by life span.

People love lists: the Ten Commandments, the Seven Deadly Sins, and the Four Beatles. But they are fascinated by rankings, which are lists organized according to some measure of value or merit. Who were the most important women in history? The best writers or most influential artists? Our least illustrious presidents? Who's bigger: John, Paul, George, or Ringo?

This is a book about measuring the “significance” of historical figures. We do not answer these questions as historians might, through a principled assessment of their individual achievements. Instead, we evaluate each person by aggregating the traces of millions of opinions in a rigorous and principled manner. We rank historical figures just as Google ranks web pages, by integrating a diverse set of measurements about their reputation into a single consensus value.

Significance is related to fame but measures something different. Forgotten U.S. president Chester A. Arthur (1829–1886) [499] is more historically significant than young pop singer Justin Bieber (1994–) [8633], even though he may have a less devoted following and lower contemporary name recognition. Significance is the result of social and cultural forces acting on the mass of an individual's achievement. We think you will be impressed by the extent to which our results capture what you think of as “historical significance.” And our computational, data-centric analysis provides new ways to understand and interpret the past.