Trace and evaluate the argument in a letter: sound reasoning, relevant and sufficient evidence

Learning goal: Trace and evaluate the argument and specific claims in a text, assessing whether the reasoning is sound and the evidence is relevant and sufficient to support the claims, and cite several pieces of textual evidence for each assessment.

Original letter — Give the Land Back. A letter to the invented Harlow Junction Courier by Walt Brenner, the invented farmer named in the companion article. It responds to the same hypothetical engineer's report; every figure, relative and story in it is fictional.

My family has farmed beside the Kettle Creek Branch since my great-grandfather sold the railroad the strip in 1884. Since the last train in 2019 the only traffic on it has been deer and the occasional four-wheeler. Now a coalition from town wants the county to spend millions turning it into a trail past my barn. The commission should let the railroad pull up its rails and give the land back to the people who farm beside it.

Start with what the grant does not pay for. The coalition says the state will cover 75 percent. True — of construction. The engineer's report puts upkeep at $61,200 a year, and a cost that comes back every year has to be compared with other costs that come back every year, not with a grant that arrives once. Sixty-one thousand dollars is more than the county spends each year on roadside weed control. And that is before anything breaks.

The county's record with money is not reassuring. Last winter the road department ran out of salt by the second week of February. Are these the people we want maintaining a bridge?

The coalition says its 220 members will do the upkeep themselves. My cousin farms along a trail two counties over. In its first summer his gate was left open twice and a heifer ended up on the county road. Volunteers were going to fix that too. Pledges are not budgets, and a pledge made in a warm meeting hall is not a mower on a July afternoon.

And who would use it? Nobody rides a bicycle past six miles of corn. The town people who filled out the county's survey do not live along the line; the people who do live along it signed my petition, nineteen of thirty-one. A survey of people who will never see the trail from their kitchen window counts for nothing.

Consider, too, where this ends. Once the county owns a strip through my farm, it will find other uses for it: a utility line, a drainage project, a road. A corridor "held" is a corridor waiting for the next idea.

I want to be clear about what I am not saying. I am not against trails; I have walked Pine Run myself. Unlike the coalition, though, I would have to live beside this one, and unlike the engineer, I do not get paid to inspect it. The engineer says the county should hold the corridor because it cannot be put back together. I say it was never the county's to begin with. The deeds are plain: when the trains stop, the land comes home.

Pull up the rails, take down the bridge and give the land back.

Tracing the argument. To trace an argument is to follow it from its main claim through each reason to the evidence offered for that reason, in the order the text gives them, before judging any of it. Judging an argument nobody has traced is judging an impression. The letter's main claim is stated at the end of the first paragraph and repeated in the last line: the commission should let the corridor revert to the adjoining owners. It offers six reasons. Reason 1, cost: the grant pays construction only, and $61,200 a year recurs; evidence is the engineer's figure and a comparison with the county's weed-control spending. Reason 2, the county's competence: evidence is the road department running out of salt. Reason 3, volunteer pledges fail: evidence is a cousin's story of an open gate and a heifer. Reason 4, nobody would use it, and the survey does not count: evidence is the claim about bicycles and corn, and the petition of nineteen. Reason 5, county ownership leads to other uses: no evidence, only a prediction. Reason 6, the land was never the county's: evidence is "The deeds are plain." Laying the argument out this way already shows that the reasons are of very different kinds, and the evaluation must take them one at a time.

Assessing whether the reasoning is sound. Reasoning is sound when the conclusion follows from the reasons and the reasons are true or well supported. Both conditions are needed. Reason 1 contains the letter's soundest reasoning: "a cost that comes back every year has to be compared with other costs that come back every year, not with a grant that arrives once." The sound core of that sentence is that a one-time construction grant does not pay for yearly upkeep, so the two can only be compared over a common span of years; that is exactly what the article does with its twenty-year totals, and the conclusion that the grant does not settle the cost question follows. Taken literally, "not with a grant that arrives once" goes too far, since the article compares them by converting both to twenty-year figures. Evaluation is not fault-finding; a fair reader credits this before turning to weaker reasons. Reason 4 contains the letter's least sound reasoning: from "the town people who filled out the county's survey do not live along the line" he concludes that the survey "counts for nothing." The premise is unverified: the article gives no breakdown of where the 900 respondents live, and some adjoining owners may have answered the survey themselves. Even if the premise were true, the conclusion would not follow, because a county decision affects the county, and the article's own point is that the survey and the petition measure different groups, not that either is worthless. Reason 5 is an unsupported prediction: it says that once the county owns the strip it "will find other uses for it" and lists three (a utility line, a drainage project, a road) without evidence that any has been proposed or that the county could not refuse them. The letter does not say that one use leads to the next. Its "where this ends" has the form of a slippery slope, in which a first step is treated as leading on by itself, but a list of possibilities is not evidence that any of them will happen. Reason 6 rests on "The deeds are plain," but the article reports the county's attorney calling them "less plain than either side would like," so the premise is contested, and reasoning from a contested premise as if it were settled leaves the argument's soundness unestablished: the premise is not shown to be false, but it has not been shown to be true either.

Assessing whether the evidence is relevant. Evidence is relevant when it bears on the claim it is offered for. The test is a question: if this piece of evidence were the opposite, would the claim be more or less likely? Ask it first about the claim the evidence sits directly under, and only then about the main claim. Apply it to the road salt. Its immediate claim is Reason 2, that the county's record with money is not reassuring. If the road department had ended the winter with salt to spare, would that reason be a little weaker? Yes, so the salt shortage is relevant to Reason 2, though one winter in one department is far from sufficient for it, and a hard winter can empty a salt shed without any mismanagement. Now ask about the main claim: would salt to spare make the corridor more or less worth returning to the landowners? Hardly at all. The letter tries to connect the two with "Are these the people we want maintaining a bridge?", but that step is a leap, since road salt and bridge upkeep are different jobs. So the fact reaches the main claim only weakly and indirectly, and a reader should notice the transition that brought it in, "The county's record with money is not reassuring," which changes the subject while keeping the tone of argument. Now apply the test to the cousin's heifer. If the cousin's gate had never been left open, would the claim that volunteers fail to prevent problems be weaker? Slightly, yes. The story is relevant. Relevance is a relationship between a piece of evidence and a particular claim, which is why tracing comes first.

Recognizing irrelevant evidence when it is introduced. Some statements in an argument fail the test for every claim they could support, and a reader should name them as irrelevant at the point where they enter. The letter's seventh paragraph says, "unlike the engineer, I do not get paid to inspect it." Apply the test to the engineer's reason, that a divided corridor cannot be put back together: if she inspected the trestle for nothing, would that reason be any more or less true? No. Would the corridor be any more or less worth returning? No. The remark is about the person, not about her reason, and it bears on neither the reason nor the main claim; it works on readers by hinting that she gains from holding the corridor, which the letter never shows. The same is true of "I have walked Pine Run myself": it supports his statement that he is not against trails, but it is not evidence for reversion. Two signals help a reader catch irrelevant evidence as it arrives: a transition that changes the subject ("unlike the engineer..."), and a statement about who someone is where the argument needs a statement about what is true.

Assessing whether the evidence is sufficient. Evidence is sufficient when there is enough of it, of the right kind, to justify the strength of the claim. Relevance and sufficiency are separate tests. The heifer story is relevant but not sufficient: one farm, one summer, two open gates, reported second hand, cannot support the general claim that pledges fail everywhere, though it does support the modest caution that pledges need a budget behind them, which the coalition chair herself concedes in the companion speech. "Nobody rides a bicycle past six miles of corn" is offered with no evidence at all, and evidence that bears on it exists in the article: Pine Run Trail's counter recorded 38,000 passes in its third year in a neighboring county. The letter does not mention that figure, and a reader who knows it should assess the claim as not merely unsupported but weakened by available evidence: the counter records passes rather than riders, and the article does not describe what Pine Run passes through, but 38,000 passes in a year in the next county is hard to square with "nobody." The petition, nineteen of thirty-one, is relevant and sufficient for the claim that most adjoining owners oppose the trail; it is not sufficient for the larger claim the letter makes with it, that the county should therefore choose reversion, because a petition measures one group's opinion on one question. Even if all thirty-one owners had signed, their opposition would not settle the costs, the deeds or the interests of the county's other residents. The weed-control comparison is relevant, since it translates $61,200 into something a reader can picture, but it is given without a figure, so its sufficiency cannot be checked; note it as unverified rather than false.

Evaluating the whole. The letter's strongest parts are the recurring-cost principle, which is sound, and the petition, which is relevant and sufficient for a limited claim. Its weakest parts are the salt shortage, which reaches the main claim only indirectly, the irrelevant remark about the engineer's pay, the unsupported claim about corn, the overreach from "different people" to "counts for nothing," the unsupported prediction of other uses and the contested deeds. A fair evaluation says both and orders them. It also sets aside rhetoric such as "a mower on a July afternoon" as vivid phrasing rather than evidence: it makes the reader picture something, which is persuasion, and it supplies no support, which is what evaluation measures.

Worked trace: Main claim: let the corridor revert. Reason 1: upkeep recurs and the grant does not cover it. Evidence: $61,200 a year from the report; weed-control comparison. Reason 2: the county cannot be trusted with money. Evidence: salt ran out in February. Reason 3: volunteers fail. Evidence: cousin's gate and heifer. Reason 4: nobody will use it and the survey is worthless. Evidence: assertion about corn; petition of nineteen. Reason 5: county ownership invites other uses. Evidence: none; a prediction. Reason 6: the land is owed back. Evidence: "The deeds are plain." Traced this way, Reasons 2 and 5 are already visible as the thinnest: one rests on a single winter in another department, the other on no evidence at all.

Worked relevance test: Take the salt shortage. Suppose the opposite: the road department finished the winter with salt to spare. First ask about its immediate claim, Reason 2: is the county's record with money now a little more reassuring? Yes, so the fact is relevant there, though one winter in one department is far from sufficient. Then ask about the main claim: is the corridor now more or less worth returning to the landowners? Barely, if at all. The main claim hardly moves, so the evidence reaches it only weakly, through a reason the letter has not established. Attach the same fact to a different claim, "the county should have an outside engineer check its cost estimates," and ask again: does a budgeting failure bear on whether estimates need checking? Yes. The fact is unchanged; its relevance changed with the claim. Now take "unlike the engineer, I do not get paid to inspect it" and run the test on every claim in the trace. None moves. That is irrelevant evidence, and the place to say so is the sentence where it is introduced.

Worked sufficiency assessment: Claim: nobody would use the trail. Evidence offered: "Nobody rides a bicycle past six miles of corn." No source, no count, no comparison. Available evidence not offered: the article's Pine Run counter, 38,000 passes in the third year, in a neighboring county. Assessment: the claim is unsupported by the letter and weakened by evidence the writer could have addressed. Compare Reason 1, where one well-sourced figure, $61,200, is sufficient for the modest claim that the recurring cost is real; sufficiency depends on how strong the claim is.

Argument tree with verdicts: Draw the letter as a tree: main claim at the top, six reasons beneath, evidence beneath each reason. Beside each piece of evidence write three verdicts: relevant to this reason or not, sufficient for this reason or not, sourced or not. Beside each reason write whether the reasoning from evidence to reason is sound, with the test you applied (opposite-evidence test, does-the-conclusion-follow test, contested-premise check). Circle any evidence that is irrelevant or only weakly relevant and write the transition that introduced it and the different claim it would support.

Independent performance: Write a four-paragraph evaluation of the letter. Paragraph one traces the argument in your own words. Paragraph two identifies the soundest reasoning, explains why it is sound, and cites the sentence. Paragraph three identifies one piece of irrelevant or weakly relevant evidence and one insufficient piece, explains the test that exposes each, and states what claim the weakly relevant evidence would better support. Paragraph four assesses the survey reasoning and the deeds premise and gives an overall judgment that credits the argument's strengths. Success criteria: every verdict cites the sentence it judges; relevance and sufficiency are assessed separately; the corn claim is compared with the article's Pine Run figure; and the evaluation names at least one strength before any weakness.