Reviews
Part of Political speeches you can actually verify
Ranking political speeches vs grouping them by the problem
Best political speeches lists measure survival, recording and curriculum. What ranking hides, and a grouping by the speaker's problem that works better.
What to take away
- Group political speeches by the problem the speaker faced, not by rank.
- Ranking puts unlike occasions on one scale, and no such scale exists.
- A usable entry names the speaker, the date, the venue and the audience.
- Write the inclusion rule before you look at candidates.
- Four speeches facing one problem compare well. Twenty-five facing twenty-five do not.
What a ranked list is really measuring
A ranked list of great political speeches mostly records one compiler's access. It shows what they could hear, what they could read, and what their teachers valued.
For a set of political speeches where every entry carries its own sources, political speeches shows the difference between a name and a checked entry.
Read any list as a document about selection rather than a verdict on the speeches.
Survival comes first. An entry qualifies only if something outlived the occasion: a recording, a transcript, a printed text, or a listener's notes. For most of history the last two were the only options.
The recording era splits every list. Routine capture, described in the history of sound recording, is the real boundary. From it a compiler can judge pace, pause and volume. Before it, they could only read. That is why the last century is over-represented.
Language narrows the pool. A list made in one language holds speeches in that language, plus a few with famous translations. Speeches whose force sits in the sound do not survive translation.
Curriculum and anthologies copy forward. The same names recur because they were taught and collected, then collected again. Few compilers reach primary holdings.
Example: four speeches that had to hold a divided room
Abraham Lincoln, Second Inaugural Address, March 4, 1865, United States Capitol. The war was ending and the country was not.
Franklin D. Roosevelt, first inaugural address, March 4, 1933, United States Capitol. Banks were closing, and the audience wanted to know whether the system would hold.
Nelson Mandela, inauguration address, May 10, 1994, Union Buildings, Pretoria. A negotiated transfer of power, with the old order seated in the crowd.
Barack Obama, eulogy for the Reverend Clementa Pinckney, June 26, 2015, Charleston, South Carolina. Nine worshippers had been killed at Emanuel African Methodist Episcopal Church nine days before.
Set side by side, the four differ in length, tone and structure. They share one task: hold together people who disagreed about what should happen next.
The civil rights canon is a harder test of the same method, and civil rights speeches shows why the weekly meetings that made the marches possible sit outside it.
Two more groups, with names and dates
News nobody wanted. Ronald Reagan addressed the nation from the Oval Office on January 28, 1986, hours after the Challenger shuttle broke apart.
Churchill told the House of Commons on May 13, 1940, three days into the job, that he had nothing to offer but blood, toil, tears and sweat.
Lou Gehrig called himself the luckiest man on the face of the earth at Yankee Stadium on July 4, 1939, while ending a career cut short by illness. His career record sets out the diagnosis and the date.
Leavings. George Washington's Farewell Address was published on September 19, 1796, and never delivered aloud.
Edward VIII spoke by radio on December 11, 1936, after abdicating. Richard Nixon addressed the nation from the Oval Office on August 8, 1974. Dwight Eisenhower gave his farewell on January 17, 1961.
Transcripts and dates sit in the farewell address archive kept by the American Presidency Project.
Ranking against grouping
Ranking puts a speech that steadied a frightened country beside one that won a narrow procedural vote. Nothing in those two events supports a shared scale.
Before comparing anything, rebuilding the room behind a political speech sets out what you need to know about the audience.
A ranked list
- What it assumes
- One scale of quality across every occasion
- What the number does
- Fills the bottom slots with the weakest entries
- Where it fails
- The order looks precise and is not
Grouped by problem
- What it assumes
- Speeches compare when the task matches
- What the number does
- Group size follows the evidence
- Where it fails
- Groups overlap, so one speech can sit in two
Ranking vs grouping by problem
Ranking
- Basis
- Single scale
- Comparability
- Implied, false
- Reader sees
- List of names
- Example
- 25 speeches, 25 problems
Grouping by problem
- Basis
- Shared problem
- Comparability
- Holds
- Reader sees
- Legible differences
- Example
- 4 speeches, one problem
Grouping is not more honest by nature. It is easier to check, because every entry has to answer a question you can state.
Building a grouped list that survives checking
Building a grouped list
- Write the rule before you look at candidates. Choose one thing you are selecting for, such as effect on the room.
- Define the pool and say what it leaves out. One sentence naming languages, dates and formats beats a page of justification.
- State the sort key, or say there is none. If the entries are ordered, name the order.
- Publish the near misses. Entries that almost made it explain your rule faster than the winners do.
- Mark what you could not reach. A speech read only in translation or in an anthology should be labelled.
- Disclose anything that shaped access. Rights and holdings decide what can be printed, which is why public domain status needs checking first.
