Comparing the Two Tests

Comparing the Two Tests

Both act tests share a first element and part company only on the second. Under dangerous proximity and under the Model Penal Code alike, the defendant must have acted with the specific intent to commit the target crime, and what changes is how far the defendant's conduct must have travelled beyond mere preparation. The proximity test asks whether the conduct came dangerously close to completing the offense.¹ The Code asks whether the actor purposely did something that is a substantial step in a course of conduct planned to culminate in the crime, and it counts conduct as a substantial step only when that conduct strongly corroborates the actor's criminal purpose.² The first test looks forward to what remains to be done, while the second looks back at what has already been done, so identical facts can fall short under one and suffice under the other.³

The two tests answer different questions about what attempt law is for. A proximity rule treats attempt liability as a way to prevent the completed harm, so it waits until the harm is near, which keeps attempt penalties away from remote preparation and leaves an actor who may still change course free of them.⁴ The cost lands on law enforcement, because officers who must wait until success is imminent may have to choose between arresting too early and standing by while the danger grows. Courts applying the dangerous-proximity idea feel that pressure, and in Bell v. State a court held that bringing dynamite onto the premises of the intended target was enough, because requiring more would jeopardize the safety of those who would apprehend the actor and the property under attack.⁵ The Code removes the pressure at its source. By shifting the focus to what the actor has already done, it lets police intervene earlier against dangerous persons, and it guards against overreach by excluding conduct that does not strongly corroborate purpose.⁶ The remaining risk is that people who had not firmly resolved to commit the crime will be convicted, a risk that grows when a doubtful confession leads a court to discount the substantiality requirement.⁷

The Code's approach is widely followed: roughly half the states and two-thirds of the federal circuits apply a substantial-step test comparable to it.⁸ An exam problem that does not name the jurisdiction's test therefore often has to be answered under both. Experimental work with mock jurors adds a caution about proximity instructions, because jurors were more likely to find the act requirement met under proximity wording than under substantial-step wording, possibly because phrases like dangerously close prime them to see the conduct as nearer to completion.⁹ The finding comes from an experimental setting and concerns lay jurors, and judges who police the limits of the test would not necessarily share the effect.¹⁰

Steps four and five do the real work, and the others keep them honest. Resolving intent first matters because an act test measures only conduct, so a weak intent showing defeats the charge under either test. The final step matters because an answer that reports one test's result is incomplete whenever the facts leave the jurisdiction unidentified.

Adams isolates the test as the only variable. Under dangerous proximity the inquiry runs through what remains, and three events still lie ahead of him, two of them in other people's hands. People v. Rizzo applied the same logic when four men, two of them armed, rode around looking for a payroll messenger and were arrested before they found him.¹¹ The court held that they had not attempted robbery, because they had planned the crime and were hunting for the opportunity but had not found or reached the person they meant to rob.¹² The standard it applied asks whether the acts come so near accomplishment that the crime would in all reasonable probability have been committed but for timely interference.¹³

The Code reaches Adams because its list of conduct that may be a substantial step includes possession of materials that are specially designed for unlawful use or that serve no lawful purpose of the actor under the circumstances, and a box of poisoned chocolates serves no lawful purpose.¹⁴ Adams also did more than possess the chocolates, since he bought them, poisoned them, and traveled toward the hand-off, each a step in a course of conduct planned to end in a killing and each strongly corroborating his purpose.¹⁵ The Code's list also shows that the men in Rizzo could face a closer question under it. Searching for the contemplated victim is listed conduct, while proximity cases have sometimes found the act insufficient where the defendant was still searching, so a search that strongly corroborated purpose could suffice under the Code.¹⁶

Young shows the substantial-step test at work on conduct short of any completed crime. The defendant parked behind a bank he apparently meant to rob, disguised himself, put on surgical gloves, and carried a scanner tuned to a police frequency, and the court held the evidence sufficient for attempted armed robbery after expressly choosing the substantial-step test over the proximity, probable desistance, and equivocality approaches.¹⁷ What separates Young from Rizzo is partly the test and partly the facts. Young had located and cased his target, whereas Rizzo's men had not found theirs, and reconnoitering in advance of the time for the crime probably would not suffice where physical proximity governs.¹⁸

Oscar is the case where the answer is genuinely uncertain, and the uncertainty tracks the choice of test. A proximity court that asks whether the actor has gained control of an element indispensable to the scheme will treat a filed claim as that element in an insurance-fraud plan, and Oscar has not filed one.¹⁹ Under the Code, staging a theft is not one of the listed kinds of conduct, but the list does not exclude other conduct, and the question becomes whether the staging and the prepared form, read together with Oscar's statement to his sister, strongly corroborate a purpose to file a false claim.²⁰ The defense will answer that an unfiled form and a van left in a lot show only that Oscar had not yet committed himself, and a jury could accept that. A jury could as readily find the opposite, so neither result is compelled and the analysis has to say which test it applies.

The two tests do not always split. At one end, buying bullets or driving to the intended victim's house remains preparation under proximity, and at the other end, pointing a loaded gun and pulling the trigger satisfies it.²¹ The strictest formulation, the last-proximate-act test, asks whether the defendant did the final act necessary to commit the offense.²² Yet no jurisdiction requires the last act before an attempt can be charged, and the first administration of poison in a plan to kill by repeated doses is attempted murder although more doses were planned.²³ A defendant who slips the first dose into a victim's drink therefore satisfies both tests, and the defense point that further doses remained does not change that result. The published Suzy illustration marks the range between these points. Suzy, who meant to kill Bob, bought a gun, learned to use it, lured Bob to the planned scene, aimed at his heart, and pulled the trigger, and that conduct satisfies the last-act, proximity, and substantial-step tests alike.²⁴ In the variation where she stops short of firing, she has not taken the last step, but a jury could still convict under the other tests because she stalked her victim and came close to firing, and the Code is concerned with preventing harm and apprehending dangerous individuals rather than waiting to see whether the actor completes the crime.²⁵

On an exam, these comparisons dictate the shape of the answer. The most common error is announcing a single result when the jurisdiction's test is unknown, because Adams and Oscar come out differently depending on the test. A conclusion for Adams would run in this order: Adams acted with intent to kill; under dangerous proximity he has not come dangerously close, because the hand-off, the delivery, and the eating all remain, so there is no attempt; under the Model Penal Code, buying and poisoning the chocolates and going to the hand-off point are substantial steps strongly corroborating his purpose, so attempted murder is made out; the outcome therefore depends on which test the jurisdiction follows. Where the facts do name the test, the answer applies that test alone and spends its effort on the facts that matter under it.