Unit 24 · Capstone: Inquiry from the Sources

Every inquiry starts with a question, and most first questions are not good enough yet. A student writes, "Was the Pullman strike bad?" That is a yes-or-no question, and it hides the real issue: bad for whom, by what measure, compared with what? Another student writes, "What did the strikers feel in their hearts?" No source can answer that with confidence. The craft of history begins by turning a vague curiosity into a research question that evidence can actually answer.

A strong research question has three marks. It is open, so the answer is an argument and not a single word. It is answerable, so there are sources that could settle it one way or the other. And it matters: someone, somewhere, would care about the answer. "Why did rail workers across the West join a boycott over wages in one Chicago suburb?" passes all three tests. There are union records, newspapers and testimony to read. The answer is a claim that can be challenged. And the question speaks to how movements spread, which still matters.

Questions also have a scale. "Why did the United States industrialize?" needs a career to answer. "Why did the Pullman company build a town instead of just a factory?" fits in a semester. A good rule is to shrink a question until you can name the specific sources that would answer it. Then, when you have an answer, ask what larger question it helps with. Historians move between the small question and the big one constantly, and the small one is where the real work happens.

Finally, notice that the question you ask decides what counts as evidence. If you ask what the company’s managers intended, the company’s letters matter most. If you ask what daily life in Pullman was like, rent receipts, church records and a worker’s diary matter more. There is no neutral place to stand. Writing the question down, and revising it as you learn, is the first honest act of the whole project.

A primary source is evidence made by people who were there: a letter, a payroll ledger, a photograph, a court transcript, a song, a tool, a house. A secondary source is an account written later by someone who studied the primary sources: a history book, an encyclopedia entry, a documentary. The 1894 Strike Commission’s testimony is primary. A historian’s 2005 book about the strike is secondary. Both are useful. They answer different questions and need different kinds of testing.

The line moves with your question. A 1950 textbook about the Civil War is a secondary source if you are studying the Civil War. It becomes a primary source if you are studying what students in 1950 were taught to believe about the Civil War. The Chicago Tribune’s 1894 front page is a primary source for the strike and also a primary source for how newspapers of the 1890s wrote about labor. Ask yourself: primary for what?

Primary sources are not automatically more truthful. A witness can lie, forget, or see only one corner of an event. A memoir written forty years later is primary, but memory has had forty years to rearrange itself. A secondary source written by a careful scholar who read a thousand documents may be more reliable on the big picture than any single witness. What primary sources give you is closeness: the words, numbers and objects the past actually left behind, before anyone else decided what they meant.

For a local inquiry in Illinois, primary sources are closer than students expect. County recorders hold deeds going back to the 1800s. The Illinois State Archives in Springfield and the National Archives branch in Chicago hold government records. Libraries keep city directories, old maps and newspaper files. Your own family holds letters, photographs and the memories of older relatives. Learning to find these, and to say plainly which kind of source each one is, is the first skill of the work.

Historians read the bottom of a document before the top. Who wrote it? When, and how long after the event? For what audience? With what purpose? This habit is called sourcing, and it comes before reading the content, because the answers change what the content can prove. A letter written by a striker to his mother the night of a riot proves something different from a company report written for stockholders three months later, even if both describe the same street.

Chicago’s most famous legend shows why sourcing matters. For more than a century people said the Great Fire of 1871 began when Mrs. O’Leary’s cow kicked over a lantern in her barn on DeKoven Street. The barn really was where the fire started. But in 1893 a reporter named Michael Ahern admitted that he and other newspapermen had invented the cow story because it made good copy. Catherine O’Leary, a poor Irish immigrant, spent the rest of her life blamed for a disaster. In 1997 the Chicago City Council formally cleared her name. Anyone who had asked "who wrote this, and why?" in 1871 might have doubted the tale a hundred years sooner.

Purpose and audience shape even honest sources. A mayor’s speech is meant to persuade. A diary is often meant for no one but the writer, which can make it more candid and also more careless with facts. An advertisement is meant to sell. A census form is meant to count, but the questions it asks reflect what the government of that year thought worth counting. None of this makes a source useless. It tells you what kind of truth to look for in it.

A practical routine: when you open any source, write four lines at the top of your notes. Author, date, audience, purpose. If you cannot fill in a line, write "unknown" and treat the source with more caution until you can. Anonymous websites, undated photographs and quotations passed around without a citation all fail this test in the same way. They may be true, but you cannot yet know.

A newspaper in July 1894 called the strikers a mob. To judge that word you need context: what was happening around the writer. The country was in the second year of a deep depression. Unemployed men had marched on Washington that spring. Only eight years earlier, in May 1886, a bomb thrown at police during a labor rally in Chicago’s Haymarket Square had killed seven officers, and four anarchists were hanged for it after a trial that Governor Altgeld later called unfair. Many middle-class readers in 1894 genuinely expected a strike to become a war. Context does not excuse the word, but it explains why it was chosen and why readers accepted it.

Contextualizing means asking what a person in that time and place could know, fear and take for granted. A letter from 1894 could not know that the strike would fail. A speech from 1919 was given in a country that had just fought a world war and was terrified of revolution. Reading a source as if its author lived today is the most common mistake beginning historians make. Reading it inside its own moment is harder and more honest.

Corroboration is the second test. One source is a claim; two independent sources that agree are the beginning of a fact. Independent is the key word. Ten newspapers that all copied the same wire report are one source, not ten. If the Tribune says cars burned on July 6 and a worker’s letter, a fire department log and the Strike Commission’s testimony all say the same, you can write that cars burned on July 6. If only the Tribune says a striker started the fire, you can write only that the Tribune claimed it.

When sources disagree, the disagreement is itself evidence. It tells you that something was contested at the time, and it sends you looking for a source with less reason to bend the truth. For the Pullman strike, historians lean on the federal commission’s report because it took sworn testimony from every side and had no payroll to protect. Even so, they read it as a document with its own purpose, written by appointed officials who wanted to prevent the next strike.

Students often say a source is "biased" as if that ends the discussion. It does not begin it. Every source has a perspective, meaning a position from which its maker saw the world. A striker’s wife and a railroad manager both saw the 1894 strike from somewhere. Having a perspective is not a flaw; it is a fact about being human. Bias is narrower: a perspective that leads a source to distort, leave out or invent, usually because the maker had something to gain or protect.

The useful question is not "Is this source biased?" but "For what is this source reliable?" A company’s newsletter is a poor guide to what workers thought. It is an excellent guide to what the company wanted workers to think, and to what the company chose to boast about. A political cartoon is unreliable about facts and very reliable about what its readers already believed. Reliability is always reliability for a purpose.

Consider the report on Chicago’s race riot of 1919. That summer a Black teenager named Eugene Williams drowned after being struck by a stone while swimming near a beach that whites claimed as their own. Days of violence followed, and thirty-eight people died, twenty-three of them Black. The governor appointed a commission of six Black and six white members to study the causes. Its 1922 report is reliable about housing, jobs and police conduct because it gathered testimony from every side and published its numbers. It is less reliable about what the commission’s own members privately thought, because it was written to be accepted by both races.

Perspective also tells you what is missing. A source written by the people in power rarely records the names of the people they ruled. Slave owners’ ledgers list human beings by first name and price. To hear the enslaved, historians turn to different evidence: interviews, court cases, songs, runaway advertisements that describe a person’s scars and skills in the owner’s own words. The historian’s job is not to find the one unbiased source, which does not exist, but to gather enough perspectives that the distortions of each can be seen.

A number feels like the end of an argument. It is usually the start of one. Between 1910 and 1920 the Black population of Chicago grew from roughly forty-four thousand to roughly one hundred nine thousand, according to the census. That is a fact, and it is also a question. Who moved, from where, and why? The number cannot say. Historians pair it with letters to the Chicago Defender, train schedules from Mississippi, and wage tables from the stockyards to explain the Great Migration. The statistic points; the other sources explain.

Every number was counted by someone with a method. The census counts people where they sleep on one night, so it misses people who move often and people who avoid the government. Studies of the census have found that Black Americans and the poor were undercounted in many years. Crowd sizes at strikes and marches were guessed by police, organizers and reporters, each with a reason to guess high or low. Before you trust a figure, source it as you would a letter: who counted, how, and why.

Definitions matter as much as counting. Unemployment can be measured as people without any job or as people looking for one, and the two figures differ by millions. A "riot death" may include people killed by police or only people killed by rioters. When two sources give different numbers for the same thing, look first at how each defined the thing. Often both are right about different questions.

Finally, beware of the leap from together to because. When two things rise at the same time, the numbers alone cannot tell you that one caused the other. Chicago’s population and its number of saloons both grew in the 1880s; neither caused the other. A historian uses numbers to describe patterns and to test claims, then reaches for other evidence, such as what people said they were doing and why, to explain them. A chart is a piece of evidence, not a verdict.

Most people in the past left no papers. They did not write memoirs or keep diaries, and no newspaper quoted them. Oral history, the recorded interview, is one of the few ways to reach them. In the late 1930s the federal Writers’ Project sent interviewers across the South to record the memories of more than two thousand elderly people who had been enslaved as children. In Chicago, the radio host Studs Terkel spent decades recording ordinary people talking about their work, the Depression and the Second World War. Without these recordings, whole lives would be known only from the ledgers of the people who ruled them.

Memory is a source with its own rules. People remember feelings more accurately than dates. They fold later knowledge into earlier events, so a person may "remember" a detail they actually read years afterward. They shape stories to make sense of their lives, which is human and also a warning. The formerly enslaved people interviewed in the 1930s were often speaking to white interviewers in the segregated South, and some plainly said less than they knew. Sourcing applies: who was asking, where, and what did the speaker have to fear?

Good interviewing is a craft. Ask open questions, not leading ones: "What do you remember about the plant closing?" rather than "Were you angry when the plant closed?" Let silence sit; people fill it. Ask for specifics: a street, a name, a price, a season. Record with permission, and note the date and place. Afterward, corroborate what you can. If your grandmother says the family came to Chicago in 1952 after the mill shut, a city directory, a church record or a census sheet may confirm the year, and the mill’s closing can be checked in the newspaper.

Oral history does something no other source does. It lets a person explain their own life in their own words, including what they thought it meant. A deed tells you a family bought a house on the South Side in 1948. Only an interview tells you what it felt like to be the first Black family on the block, or the white family that sold and left, or the child who did not understand why the neighbors stopped waving. That kind of evidence is fragile and irreplaceable. Every year it becomes less available.

Research can hurt people. From 1932 to 1972 the U.S. Public Health Service studied hundreds of Black men in Alabama who had syphilis, told them they were being treated, and withheld the cure even after penicillin became standard in the 1940s. When a reporter exposed the Tuskegee study in 1972, the outrage led to new federal rules. A 1979 report laid out principles that still govern research on human beings: respect for persons, which means informed consent; doing no harm; and fairness in who bears the risks. Universities now require a review board to approve studies involving living people.

A student historian works at a smaller scale, but the same duties apply. Before an interview, explain what the project is, how the recording will be used, and who might hear it. Let the person refuse any question and stop at any time. If a relative tells you something that could embarrass or endanger a living person, ask before you use it. Records that name private people, such as court files, medical notes or a neighbor’s mortgage, deserve extra care even when they are legally public. The question is not only "Can I use this?" but "Should I, and how?"

Honesty is the second duty. Never invent a quotation, a number or a source, and never trim a quotation so that it says the opposite of what the speaker meant. Report the evidence that hurts your argument as faithfully as the evidence that helps it. A historian who hides a contradicting source is not making a stronger case; she is making a weaker one that will collapse the first time a reader checks. Your notes should let another person retrace every step you took.

The third duty is credit. Plagiarism, using someone’s words or ideas as if they were your own, is theft of the one thing a scholar owns. It is also a failure of evidence, because a reader cannot check a source you have hidden. Quote when the exact words matter, paraphrase when they do not, and cite in both cases. This includes ideas you found in a secondary source, and it includes text produced by a computer program, which cannot be a source and must never be passed off as your reading of one.

A claim is a sentence someone could reasonably disagree with. "The Pullman strike happened in 1894" is not a claim; it is a fact. "The Pullman strike failed mainly because the federal government sided with the railroads" is a claim, because a thoughtful person could argue instead that it failed because the union overreached or because public opinion turned. A good claim answers your research question directly and can be stated in one sentence you would be willing to defend out loud.

Evidence is the specific material from your sources that bears on the claim: a quotation, a number, a date, a description of a document. It must be specific. "Newspapers were against the strike" is not evidence; "the Tribune’s front page of July 7 described the strikers as a mob and called for troops" is. Choose the strongest evidence, not the most, and always say where it came from so a reader can check.

Reasoning is the part students most often skip. It is the sentence or two that explains why the evidence supports the claim, in your own words. Evidence never speaks for itself. If your evidence is that Cleveland sent troops over Altgeld’s objection, your reasoning must say why that shows the government sided with the railroads rather than, say, protecting the mail. Reasoning is where you show the reader your thinking, and where a careful reader will test it.

The last move is the counterclaim: the best argument against your own. State it fairly, not as a straw man, then answer it with evidence. If the counterclaim is that the union’s boycott was illegal and the government simply enforced the law, you might answer that the injunction the government used was itself new and controversial, and that the commission blamed the company first. A claim that has survived its strongest opponent is worth more than a claim that was never tested. Structure this way and a reader can follow every step, agree or disagree, and know exactly where.

A citation is a set of directions. It tells a reader how to find the exact source you used: who made it, what it is, where it lives, and when. Historians most often use footnotes in the style of the Chicago Manual of Style, first published by the University of Chicago Press in 1906, though other styles do the same job. The style matters less than the promise behind it: anyone who doubts you can go and look.

Different sources need different directions. A book needs author, title, publisher, place and year, plus the page. A newspaper needs the paper’s name, the date and the page or headline. A census table needs the year, the table or volume, and where you found it. A letter in an archive needs the writer, the recipient, the date, the name of the collection and the archive that holds it. An interview needs the person’s name, the interviewer, the date and place, and where the recording is kept. A website needs the title, the author if known, the address and the date you looked at it, since web pages change.

Citation also protects you. When a reader asks "how do you know that?", the footnote answers. When two sources disagree, the citations let the reader see which you chose and why. And when you return to a project months later, your own citations are the only reason you can find that one letter again. Keep a running list from the first day; reconstructing sources at the end is where most mistakes and most lost weekends come from.

A last point. Cite the source you actually used. If you learned about the Strike Commission’s finding from a historian’s book, cite the book, or find the report and cite that. Citing a document you have never seen, because someone else quoted it, passes along their errors as your own. Historians call this a secondhand citation, and a careful reader can usually spot one.

In the early 1900s the leading historians of Reconstruction, the years after the Civil War, taught that giving Black men the vote had been a mistake and that Southern white "redeemers" had rescued the region from corrupt rule. Textbooks repeated this for decades. In 1935 W. E. B. Du Bois published a very different account, arguing that Reconstruction governments had built the South’s first public schools and that Black voters and officials had been slandered by the people who overthrew them. He was largely ignored by the profession for years. By the 1960s and after, historians working with the same records reached conclusions much closer to his. Today his view is the mainstream one.

What changed was not the past. What changed was the evidence historians chose to read, the questions they asked, and who was doing the asking. The early historians read the papers of white planters and politicians. Later historians read the records of the Freedmen’s Bureau, Black newspapers, and the testimony of freedpeople before Congress, which had been sitting in the archives all along. The civil rights movement made the question of Black citizenship urgent, and more Black scholars entered the field. New questions found evidence that old questions had walked past.

This is the ordinary life of history. A generation asks about presidents and generals; the next asks about workers, women and the enslaved; the next asks about the environment or about ordinary consumers. Each generation finds sources the last one did not need. Disagreement does not mean that anything goes. Every interpretation must still answer to the evidence, and some interpretations lose because the evidence turns against them, as the early Reconstruction story did. Historiography, the study of how historians’ accounts have changed, is a record of arguments won and lost.

For your own work this has a practical meaning. When two secondary sources disagree, do not average them. Ask when each was written, what sources each used, what question each was answering, and what the author’s own moment made visible or invisible. Then go to the primary sources yourself. You are not the last word either. You are joining a conversation that started before you and will continue after, and the best thing you can add is a claim honestly built from evidence someone else can check.

Historians do not begin with the whole past. They begin with one door. For a student, the four easiest doors to open are a place name, a school, a family and a law. Each one is small enough to finish in a semester and each one connects to something much bigger. A street name in Chicago can lead to a civil rights campaign, a real estate deal or a war. A school’s name can lead to the person it honors and to the neighborhood that fought to build it. A family’s move can lead to the Great Migration, to a war in another country or to a factory that closed. A law can lead to the people it helped and the people it hurt.

Take the street door. Chicago has thousands of named streets, and every name was chosen by somebody at some moment for some reason. In 1909 the city renumbered nearly all of its addresses so that State Street and Madison Street became the zero point of a single grid, and in the years around it many duplicate street names were changed. That means a street’s name and its numbers may date from different decades. A local question about a street might be: who chose this name, when, and what did people say about it at the time? Ida B. Wells Drive answers all three, because the City Council debated it in public in 2018 and the debate was recorded.

Take the family door. A student whose grandparents came to Chicago from Mississippi, from Mexico, from Poland or from Puerto Rico holds a primary source at the dinner table. The question is not simply where did we come from, which is a fact, but why did they leave then, why did they choose here, and what did they find when they arrived. Those are questions of cause and consequence, and they can be answered with interviews, letters, photographs, census pages and the records of the employers and churches the family joined. The answers usually surprise the family itself.

Take the law door. A law is a decision written down, so its trail is unusually clear: the text, the votes, the court cases that tested it and the numbers that changed after it. A student might ask what a 1940 Supreme Court case brought by a Chicago family named Hansberry did to the private agreements that kept Black families out of white neighborhoods. Carl Hansberry, whose daughter Lorraine later wrote A Raisin in the Sun, won his case on narrow grounds. Those agreements stayed common until a broader ruling in 1948. That gap between winning in court and winning on the street is a question about a law’s effect. It is one of the most useful questions a citizen can learn to ask.

A question can be interesting and still be a bad project, because no records survive to answer it. Before committing, do a feasibility check. Ask four things. Do sources exist? Can I reach them? Can I read them? Can I finish in the time I have? A student who wants to know what a great-grandmother thought about the 1919 riot, when the great-grandmother left no letters and everyone who knew her has died, has a moving question and no evidence. A student who asks instead what the neighborhood’s newspapers said about the riot has a question the Chicago Public Library can answer this week.

The first test is existence. Some records were never made, some were made and thrown away, and some were destroyed on purpose. Fires took much of Chicago’s early paper in 1871, and the 1890 federal census was largely lost to fire and water damage in Washington decades later. Records of the poor, of the enslaved, of women and of children were kept less often than records of property owners. When the record you want does not exist, you look for the records that were kept about the same people by someone else: a tax roll, a court file, a church register, a company payroll. This is called reading against the grain, finding the missing person in someone else’s paperwork.

The second test is reach. Some records are online, some are in a reading room downtown, some are in a county building in another state, and some are private. A church may share its records with a polite letter; a company may not. Government records in Illinois are generally open under the state’s Freedom of Information Act, which requires a public body to answer a written request within five business days, though it can take longer to get the actual documents. Family papers belong to the family. Part of a feasibility check is simply asking who holds the record and whether they will let you see it.

The third and fourth tests are reading and time. A record in Polish or Spanish is not a problem if you or a partner can read it; a record in nineteenth-century handwriting takes practice; a record in Latin may be beyond reach. Time is the test students most often skip. A question that needs three archives and twenty interviews is a book, not a semester project. Narrow it. Instead of the history of the school, ask how the school got its name. Instead of the whole Great Migration, ask about one family in one decade. A narrow question answered well beats a grand question abandoned in March.

Return to the student on the bus. Her first question was simple: who was Ida B. Wells? Her second was better: why did the city rename a street for her in 2018 rather than in 1931, when she died in Chicago, or in 1962, a century after her birth? To answer it she had to build the timeline of Wells’s life from the records and then build a second timeline of how Chicago remembered her. Those two timelines did not match, and the gap between them became her project.

The first timeline came from Wells herself and from the courts. In her autobiography, written late in life and published after her death, she describes buying a first-class ticket in Tennessee in 1883, refusing to leave the ladies’ car and being dragged off the train. The court records confirm that she sued and that a Memphis court awarded her five hundred dollars in 1884. In 1887 the Tennessee Supreme Court reversed the ruling. Five years later, in 1892, three of her friends who ran a grocery store in Memphis were taken from jail by a mob and murdered. Wells investigated lynchings across the South, published her findings in a pamphlet called Southern Horrors, and while she was away a mob destroyed her newspaper office. She never lived in the South again.

The second timeline came from Chicago’s own records. Wells settled in the city in 1895, helped found the NAACP in 1909, organized one of the first Black women’s suffrage clubs in Illinois in 1913, and ran for the state senate in 1930. After her death in 1931 the city named a public housing project for her in the early 1940s. That project was torn down in the 2000s. For most of a century the only major public thing bearing her name was a set of buildings that many Chicagoans associated with poverty, and then not even that. The push to give her a downtown street came from a campaign led by her great-granddaughter and others, and the council approved it in 2018.

The student’s claim took shape from the gap. She argued that a city’s honors show what it values at the moment it gives them, not what the honored person did. Wells was recognized only when the people asking for the honor had gained enough of a voice to be heard. She could have stopped there. But she also found the honorary street name program, the brown signs that hang beneath the real ones. Those signs had honored hundreds of people without changing any address. The 2018 vote was different because it changed the legal name. Whether that difference matters is the kind of question a good inquiry leaves for the reader to argue about.

Think of a record as a fossil. It survives where it was made. A birth was written down by a county clerk, so the record lives with the county. A house sale was written down by the recorder of deeds, so the deed lives there, and in Cook County that office is now part of the county clerk. A city council debate was written down by the city clerk, so the journal of the council’s proceedings is a public record you can read in the library or online. A student who asks who would have needed to write this down, and who paid them to keep it, will usually find the record on the first or second try.

In Chicago the main public homes for records are a short list. The Chicago Public Library’s main building downtown holds city directories, old newspapers on microfilm and the Chicago Sun-Times and Tribune in digital form. The Chicago History Museum holds photographs, maps and the papers of families and businesses. The Newberry Library holds maps, genealogy collections and the papers of writers and reformers. The Illinois State Archives in Springfield holds the state’s own records, including some court and military files. The National Archives has a regional branch in the Chicago area that holds federal court and agency records for the Midwest. Universities hold specialized collections. Each has a reading room, a catalog you can search from home, and staff whose job is to help you.

Some of the most useful records for a local project are maps and counts. Fire insurance companies drew detailed block maps of American cities, called Sanborn maps, from the 1860s into the twentieth century, showing every building, its material and its use. A student can watch a block change from wooden cottages to brick apartment houses to a parking lot by laying three Sanborn maps side by side. The federal census counts everyone every ten years, and the individual answer sheets, with names, ages, birthplaces and jobs, are released to the public seventy-two years after each count. The 1950 sheets opened in 2022. That rule protects privacy and also means the census you need for a grandparent’s childhood may already be online.

The last home for records is the internet, and it is both the best and the most dangerous. Digitized newspapers, scanned deeds and searchable census pages have made a semester project possible that would once have taken years. But the internet is a copy of an archive, not the archive itself, and copies leave things out. A search that finds nothing does not mean nothing exists; it may mean the page was never scanned or the handwriting was never read by the software. When the online trail ends, go to the building. The reading room is where the missing piece usually is.

For many local questions the most important source is a person. The woman who has run the corner store for forty years, the retired teacher who was at the school when it opened, the grandfather who was fourteen the week of the riots: their memories are evidence that exists nowhere else. But memory is fragile and interviews go wrong in predictable ways. The fix is to treat an interview as a document you are creating, with the same care you would give a document you found. Plan it, get consent, record it, transcribe it and cite it.

Plan it by writing a short list of open questions, the kind that cannot be answered yes or no. Not: Was the strike violent? But: Tell me about the day the strike began. What did you see from your window? Who was there? Open questions let the person lead, which is where the surprises come from. Avoid leading questions, which supply the answer inside the question, such as: You must have been terrified, right? People agree with leading questions to be polite, and then you have recorded your own guess in their voice. Bring a few facts and dates to help the person anchor memory, but let them correct you.

Get consent before the recorder starts. Explain who you are, what the project is, who will hear the recording, whether the person’s name will be used, and that they can stop at any time or ask that something be left out. Write this down and have them sign it, or record them agreeing to it at the start of the tape. This is not a formality. A person who tells you about a family member’s arrest or an employer’s cheating is trusting you with something that can cause harm, and the promise you make about how you will use it is a promise you must keep. Ask before you take a photograph of them or of anything they own.

Afterward, write a transcript, which is the recording turned into text with the time noted every minute or so. Transcripts let you quote exactly and let a reader check the quote against the recording. Then treat the interview like any other source. Source it: who said this, when, to whom and why? Corroborate it: does the date match the newspaper, the address match the city directory? Where memory and record disagree, do not simply pick the record. Note the disagreement and ask what the person’s memory reveals that the record cannot, such as what the event felt like or what people believed was happening. Keep the recording and the consent form together, and give a copy of both to the person you interviewed.

A law is the easiest kind of decision to research, because it is a decision that was written down on purpose and then argued about on paper for years. Its trail has four main links. First is the text: the words as passed, which in Illinois live in the state’s compiled statutes and for Chicago in the municipal code, both searchable online. Second is the legislative record: the bill’s sponsors, the committee hearings, the floor debate and the votes, which show what the people who passed it said they meant. Third is the case law: the court decisions that tested the law when someone claimed it was applied wrongly or was unconstitutional. Fourth is the effect: what changed in the world afterward, which is usually the hardest link to find.

The student who followed Ida B. Wells’s lawsuit ran straight into the third link. The Fourteenth Amendment, ratified in 1868, says that no state may deny any person the equal protection of the laws. Tennessee had passed a law requiring railroads to provide separate but equal accommodations for Black passengers. When Wells sued, the trial court found that the smoking car was not equal to the ladies’ car and awarded her damages. The state supreme court disagreed and said, in effect, that the railroad had offered her a car that was good enough and that her real purpose had been to make trouble. Nine years later, in 1896, the U.S. Supreme Court accepted the separate but equal idea in Plessy v. Ferguson. Reading those two decisions together, the student could see the legal reasoning that would hold for another sixty years.

The fourth link, effect, takes a different kind of evidence. A law that says a thing does not make the thing happen. To measure effect you compare before and after, and you compare places with the law to places without it. Did a fair housing law change who lived where? Look at the census tracts before and after. Did a minimum wage law raise pay? Look at wage surveys and at the businesses that opened and closed. Did a school funding formula help poor districts? Look at spending per student by district over time. These are questions of causation, and the rule from the last chapter still applies: a change that follows a law is not proof the law caused it, because other things changed at the same time.

Two cautions. The law’s text and its enforcement are different things, so the record of who was actually arrested, fined or sued under it may tell a different story from the words. And laws that harmed people were often written in neutral language. The Chicago covenants that Carl Hansberry challenged never mentioned race in their titles, and the 1909 renumbering ordinance said nothing about who lived on the renumbered streets. Reading a law’s effect means reading past its language to the people it touched, which is why a law project nearly always ends in an interview or a census page.

By now the student has a pile: a court decision, an autobiography, three newspaper articles, a council journal, two interviews and a map. The pile is not an argument. The step between them is a source table, a single page with one row per source and columns for who made it, when, for whom, what it says about the question, what it cannot say, and how it agrees or disagrees with the others. Building the table takes an afternoon and it does two things. It shows where the evidence is thick and where it is thin, and it forces every source to face every other source.

The claim comes out of the table, not before it. Students often write the claim first and then hunt for evidence to decorate it, which is how bad history and bad politics both work. Instead read down the columns and ask what pattern the sources make. The Wells student saw that every honor the city gave came after a campaign by Black Chicagoans who led organizations that officials had to answer to. Her first draft claim was: Chicago honored Wells when Black Chicagoans gained power. That was a start, but the table pushed back. The council records showed that the campaign’s first proposal, in 2017, had been to rename Balbo Drive, a downtown street named in the 1930s for an Italian aviator who was a leader in Mussolini’s government, and that this proposal stalled after Italian American groups objected. The claim had to grow more precise.

A claim should be debatable, specific and supportable. Debatable means a reasonable person could argue the other side; a claim nobody would deny, such as Wells was important, is a fact, not a claim. Specific means it names who, what, when and how, so it can be tested. Supportable means the sources in your table actually bear on it. The student’s revised claim read: Chicago’s honors for Wells came only when organized Black Chicagoans with access to the council made the demand, and even then only when the honor cost little; the 2018 renaming succeeded because it replaced a name almost no one defended. That claim can be argued with, and it can be checked.

Wells herself modeled this. Her pamphlets on lynching did not rest on outrage alone. She gathered the counts of lynchings published by white newspapers, the stated reasons given by the mobs, and the details of individual cases, and she showed from those sources that the mobs’ own explanations did not hold. Her method was a source table before the term existed. The lesson for a student is that a claim which comes out of the evidence is far harder to knock down than a claim the evidence was recruited to serve.

Every claim worth making has a counterclaim, the best case a thoughtful person could make against it. Weak arguments ignore the counterclaim or mock a feeble version of it. Strong arguments state the counterclaim so fairly that its own supporters would nod, and then answer it. The habit has a name in debate: steelmanning, building the strongest possible version of the other side, the opposite of the straw man that is built to be knocked down. It is also simply honesty, because if you cannot state the other side’s case you do not understand your own.

The Wells student’s counterclaim came from a council member’s remarks in the 2018 debate. The member argued that the renaming was not about power at all but about time. The city honors people when history has settled their reputation, and Wells’s reputation grew as historians rediscovered her in the 1970s and after. That is a real argument with real evidence behind it, because scholarly books about Wells did multiply after 1970. The student had to concede the part that was true: national attention to Wells did rise, and that rise made the renaming easier. Then she had to rebut the part that was not, with evidence. The 1940s housing project showed the city already knew who Wells was. The stalled 2017 proposal for Balbo Drive showed that reputation alone did not move the council when the honor had a cost. The 2018 campaign’s own records showed organized pressure, not a quiet consensus.

Notice the shape of that paragraph. Counterclaim stated fairly. Concession of what is true. Rebuttal with specific evidence, each piece tied back to the claim. That structure works for any local question. If your claim is that a school was named to honor a neighborhood’s demand, the counterclaim may be that the board simply followed a naming rule. If your claim is that a family left the South because of violence, the counterclaim may be that they left for wages. Find the evidence for the other side before someone else does, and put it in your own paper.

A famous model of answering a majority is the dissent. When the Supreme Court approved separate but equal in Plessy v. Ferguson in 1896, Justice John Marshall Harlan wrote alone that the decision would prove as harmful as the worst rulings of the past. He did not pretend the majority had no argument; he restated it, then showed what it would produce. His dissent lost in 1896 and won in 1954, when the Court reversed Plessy in Brown v. Board of Education. A good rebuttal is written for readers who are not yet born.

An inquiry that lives only in a folder has not finished its work. The last research step is to put it in front of people who did not do it, and the first question is who those people are. A class of peers, a school board, a room of neighbors at a library, the family whose story it tells and the internet each need a different form. A written paper carries the most evidence. An exhibit of maps and photographs on a wall reaches people who will not read ten pages. A ten-minute talk with slides can move a room. A short audio piece built from the interviews lets the people in the story speak for themselves. Choose the form for the audience, not for the grade.

Whatever the form, three things must be visible within the first minute or the first page: the question, the claim and the reason anyone should care. Then the evidence, chosen rather than dumped. A presentation is not the source table read aloud. Pick the two or three pieces that carry the most weight, show them directly, a photograph on the screen, a line from the court decision read out loud, thirty seconds of an interview, and explain in your own words why each one supports the claim. Then give the counterclaim its own moment. An audience that hears you state the other side fairly will trust everything else you say.

Visual evidence needs the same sourcing as written evidence. A map on a slide should carry its date and its maker. A photograph should say who took it, when and where, and if you cropped it, say so. A chart should name its source and its unit, and its vertical axis should start at zero unless you tell the audience why it does not. These are not decorations; they are the citations of a visual argument. An audience of neighbors will forgive an ugly slide and will not forgive a misleading one.

Plan for questions, because the questions are where an inquiry becomes a conversation. Bring the source table so you can answer where did you find that. Practice saying I do not know, and saying it without apology, because a researcher who admits the limit of the evidence is more believable than one who pretends there is none. If someone in the room lived the history you are presenting, and in a local project someone often did, their correction is not an embarrassment. It is a new source. Write it down and thank them by name.

The Illinois social science standards end every inquiry with the same step: taking informed action. The word that matters is informed. Anyone can be angry about a street name or a school’s closing; a researcher can say exactly what the records show, what they do not, and what a specific decision-maker could do about it. The action can be small. A letter, a two-minute statement at a public meeting, a written comment on a proposed rule, a proposal to a principal or a library. What makes it informed is that the claim, the evidence and the counterclaim are all in it, and what makes it action is that it is addressed to someone who can act.

Start by finding the jurisdiction, the body that actually has the power over your question. A street’s legal name in Chicago is changed by the City Council. The person to write is the alderperson for that ward and the council committee that handles it. A school’s name is decided by the school district’s board. A state law is the business of your state representative and state senator in Springfield. A federal rule is the business of the agency that proposed it and of your representative in Congress. Writing to the wrong body wastes everyone’s time. Writing to the right one, with evidence, gets read, because most officials receive few letters that contain any.

Each form of action has its own rules. A public comment on a proposed federal rule is submitted during a set period, often thirty to sixty days, through the government’s rulemaking website. Every comment becomes part of the public record the agency must consider. Public comment at a Chicago City Council or school board meeting requires signing up in advance and is usually limited to a few minutes. So the statement must open with the request and follow with the two best pieces of evidence. A proposal to a principal or an alderperson should say what you want done, what it would cost, who supports it and what the records show. A letter should fit on one page.

The student who researched Ida B. Wells Drive found that the campaign’s first request had been for Balbo Drive, and that the council chose Congress Parkway partly because renaming it disrupted few addresses. Her informed action was not a demand but a proposal: that the school district add a short history of the street’s name to the curriculum of the schools along it, with her source list attached. She sent it to her local school council and to the alderperson, with the source table as an appendix. Whether it is adopted is not the measure of the action. The measure is that a decision-maker now has evidence they did not have before.

The last page of an inquiry is about the researcher. It is not a summary and not a confession; it is a record of how a mind changed under the weight of evidence, which is what historians mean when they say history is an argument and not a list. Write it in four parts. What did you believe when you started? What evidence moved you, and in which direction? What do you still not know, and why? What would you do differently if you began again tomorrow? A reflection that answers those four questions plainly is worth more than one that says the project was interesting and taught me a lot.

The Wells student’s reflection began with her first belief: that the street was renamed because Wells was a hero and the city finally noticed. She wrote that the council journal and the stalled Balbo Drive proposal broke that belief. The interviews replaced it with something less flattering to the city and more interesting. Her second belief, that honors are simply about power, was softened by the council member’s counterclaim and by the scholarly books. Her final claim held both. That is revision in the true sense: not fixing spelling, but re-seeing.

She was also honest about the holes. She never found the campaign’s own internal letters and had to rely on one interview for how it chose its targets. She could not read the Tennessee court’s full file, only the published opinion. She did not interview anyone who opposed the renaming, and she said so, because a source set with no opponents in it is a warning sign. Naming the limits of your evidence is not weakness; it tells the next researcher where to dig, and it tells the reader exactly how far to trust you. Wells did the same in her pamphlets, noting which cases she had confirmed and which rested on a single report.

Finally, the reflection should look forward. The student wrote that next time she would start the interviews earlier, because the living sources took the longest to reach and gave the most. She wrote that she now reads a street sign, a school name and a news story differently, asking who chose those words and when. That habit is the real product of the unit. The paper will sit in a folder. The habit of asking for the evidence, of stating the other side, of saying I do not know, will go with you into every argument you have for the rest of your life, including the ones about the present.