The Truth Behind "Lies, Damned Lies, and Statistics"

The enduring quote “There are three kinds of lies: lies, damned lies, and statistics” strikes an ominous chord about the credibility of data. But the phrase itself has controversial origins wrapped in misuse and misattribution. The complicated history reveals insights into how data analytics transitioned from suspect to underpinning the modern world.

Murky Origins of 'Lies, Damned Lies, and Statistics'

The famous saying evolved over centuries, with the earliest known version appearing in the 1770 English poetry collection “The New Bath Guide”:

“Tis said, there are Lies, damn’d Lies, and the Accounts of the Corporation.”

This initial iteration targeted bureaucratic financial reporting in Bath, not broader innuendo about statistics.

Over 100 years later, the phrase took on new life when British Prime Minister Benjamin Disraeli quipped in a speech:

“There are three kinds of lies: lies, damned lies, and statistics.”

The exact year varies from source to source, but it was likely between 1871-1874 based on records of Disraeli’s speeches. His version replaced “accounts” with the modern “statistics.”

Mark Twain later popularized the saying in autobiography published posthumously in 1924. But he credited it to Disraeli, further muddying its true origins. Only recently have researchers traced the real roots back centuries before either statesman.

What did Disraeli and Twain truly mean to imply by invoking “statistics” as the worst brand of lie? The context provides clues.

Misleading Metrics

In 19th century England, statistics referred to the numerical metrics increasingly used by government agencies and newspapers when reporting data to the public.

Author Charles Dickens notably blasted the use of cherry-picked stats to inflate or deflate government performance based on political aims:

“How the Circumlocution Office may be described! How should I describe it! As a finger-post pointing out the way to Preferment, indicating clearly that it is the great road to honour, advancement, immense wealth, patronage, influence?”

Disraeli’s own Tory party engaged in public PR battles with their Liberal rivals to shape perception of the nation’s course. By labeling dubious government metrics as “statistics” not to be trusted, Disraeli sought to discredit partisan opponents and their numerical narratives.

Twain also despised the manipulation of numbers and surveys for propaganda purposes, especially when used to marginalize minority groups. The context of late 19th century official statistics helps shed light on the meaning.

Defining Data Science

But as objective, rigorous statistical analysis gained credence in business, government and academia throughout the 20th century, “statistics” steadily lost its connotation of pliability and deception. The discovery of genuine knowledge and patterns from data reclaimed the core value of statistical science.

This transition from stats as propaganda to enlightenment was captured in a 1954 address by Professor Darrell Huff, author of the seminal book How to Lie with Statistics:

“The secret language of statistics, so appealing in a fact-minded culture, is employed to sensationalize, inflate, confuse, and oversimplify,” declared Dr. Huff. “The careless statistician compares apples and oranges... Mistakes are costly. Because they are avoidable.”

Yet he concluded on an optimistic note: “There is terror in numbers... But in more honest hands, the same numbers can become instruments of integrity and progress.”

The Power of Data

With the flourishing of machine learning and predictive analytics, the vast potential of data science to combat ignorance emerged. As mathematician Eric Weinstein observed:

“Data dies in darkness. Numbers have lost their freedom to confuse and distort. Like DNA, data is often invisible to the naked eye. But today, we can remove data’s cloak of secrecy.”

The pioneers who recognized that statistical analysis could reveal true insights where deceit once lurked were ultimately vindicated. Their battle to separate truth from false narratives created the foundation of modern data literacy.

The Next Chapter

Of course, distrust in how numbers can mislead persists today and always will. As data becomes embedded in all facets of life, healthy skepticism remains necessary.

But perhaps Twain himself—who helped popularize Disraeli’s infamous phrase indicting statistics—offered the best guidance on how to interpret data with wisdom:

“Figures often beguile me,” he quipped, “particularly when I have the arranging of them myself; in which case the remark attributed to Disraeli would often apply with justice and force: ‘There are three kinds of lies: lies, damned lies, and statistics.’”

Twain’s wry acknowledgment of data’s power to inform—or deceive—is as relevant today as ever. Spreading statistical literacy and transparency provides the antidote. Because as data science continues improving lives, one truth endures: numbers may obscure facts or illuminate them. Depending on the hands they are in.

How can you ensure your organization is using data science to uncover the truth?

Work with senior experts to design the right approach to explore your data.

Contact Random Forest Services today.

British Houses of Parliament