Concept
The Five Elements
Child trespassers and attractive nuisance
The Five Elements
Section 339 of the Restatement (Second) of Torts makes a possessor of land liable for physical harm to trespassing children caused by an artificial condition on the land if five conditions, lettered (a) through (e), are all met.1 The scope limits covered above decide whether a case falls inside the section; the five clauses decide whether a case inside the section succeeds. The plaintiff must prove each clause, and a failure on any one returns the case to the traditional rule that a possessor owes a trespasser no duty of reasonable care and answers only for deliberate or reckless harm.2 A plaintiff with a vivid injury and a cheap precaution can therefore still lose on a clause that neither fact touches.
The clauses run in the order a court reasons through a case. The first three ask what the possessor should have known and what the children could understand: whether the possessor had reason to expect child trespassers at the place, whether the possessor knew of a condition carrying an unreasonable risk of death or serious bodily harm, and whether the children’s youth kept them from discovering the condition or appreciating the danger.3 The last two ask whether the possessor acted reasonably in light of those facts, first by comparing the utility of the condition and the burden of eliminating the danger with the risk to children, and then by asking whether the possessor failed to use reasonable care to eliminate the danger or otherwise protect the children.4 The weighing comes last because a possessor cannot be asked to balance a burden against a risk the possessor had no reason to anticipate.
Because one unproved clause defeats the claim, an exam answer should test each clause against the facts and spend its words where the facts put a clause in doubt, instead of reaching first for the cost of the precaution. Clause (a) comes first.
Likely Child Trespassers
Clause (a) requires that the place where the condition exists be one upon which the possessor knows or has reason to know that children are likely to trespass.5 The clause fixes on a particular place, rather than on the land as a whole, because a possessor can guard only against what the possessor can anticipate, and anticipation follows specific ground. A lot beside a schoolyard and a tract at the end of a private road may have equally careful owners, yet only the first gives its owner reason to expect an intruding child. Clause (a) is also separate from clause (b): the plaintiff must prove the possessor’s knowledge that child trespass is likely and, apart from that, the possessor’s knowledge of the condition and its danger.
The clause says “likely,” and a bare possibility that some child might wander onto the land does not meet it. Yet the possessor need not have seen a trespassing child. The possessor is charged with the facts actually known and with the logical inferences from those facts, so a known fact that makes child trespass likely supplies the reason to know.6 At the far end, a possessor who actually knows that children trespass satisfies the clause outright. In Dumbreck, many children played in the field, and the defendant’s servants warned them out while knowing that the warnings had little or no effect.7 Those facts show knowledge, yet the traditional rule gave a trespassing child no claim to reasonable care.8
The clause imposes no duty to inspect. The possessor need not patrol the land to learn whether children are coming, but cannot claim ignorance of what the facts at hand plainly show.9 A possessor who has watched schoolchildren file past the fence cannot answer that no child was ever caught on the lot.
Anderson v. Cahill shows the inference at work. A contractor left a deep excavation open at a construction site. No trespass had been reported, and the pastor of St. Mary’s Church testified that the school children had been forbidden to enter the construction area and that he knew of no child who had been there.10 The court treated those facts as evidence for the jury on whether the defendant knew or should have known that children were likely to trespass, and held that the lack of a known history of trespass cannot excuse a defendant who could reasonably have foreseen the likelihood of children’s presence at the site.11 The question turned on inference, not history: the children around the site gave a reasonable contractor grounds to expect them.
Croaker v. Mackenhausen and Fields v. Henrich show the opposite result. The court in Croaker found that the defendant had no reason to know of the child’s trespass, and the plaintiff in Fields failed to prove that the landowners knew or had reason to know that children too young to appreciate the danger were likely to trespass.12
Some courts phrase clause (a) as foreseeability or “should know” instead of “reason to know,” and Anderson itself says both “knew or should have known” and “reasonably have foreseen the likelihood.”13 The difference may be slight, because even under a general duty of reasonable care a possessor would not ordinarily investigate for trespassers without some fact suggesting the need.14 A student should state the clause in the Restatement’s words and, where a jurisdiction speaks of foreseeability, ask whether a reasonable possessor would anticipate child trespass on the facts known.
Unreasonable Risk of Serious Harm
Clause (b) turns from the place to the condition. It requires that the condition be one of which the possessor knows or has reason to know, and which the possessor realizes or should realize will involve an unreasonable risk of death or serious bodily harm to such children.16 The clause has two steps, notice that the condition exists and recognition that it is dangerous enough to matter, and both are measured on the possessor’s side of the encounter. Clause (a) asked what the possessor should expect of children at the place, and clause (c) will ask what the children could understand. The words “such children” tie clause (b) to the class that clause (a) found likely, so the risk is assessed for the children the possessor should expect to meet the condition.
On notice, the possessor must know or have reason to know both that the condition exists and that it presents an unreasonable risk of serious harm to a child of the class whose trespass is likely.17 The possessor has no duty to inspect but is charged with the facts actually known and the logical inferences from them. The limit on inspection has a reason: a duty to repair a hazard that no one knew about could not change anyone’s behavior. The charge of logical inferences guards the other flank, because it stops a possessor who has seen the signs from pleading ignorance of the conclusion they point to.
Menneti v. Evans Construction Co. shows that a condition may be foreseeably dangerous before the feature that causes the harm appears. A child drowned in a ditch after spring rains had filled it. The case turned on whether the defendants “maintained” the condition, which the district court defined as an intentional retention of an artificial condition after the possessor had actual or constructive knowledge of its existence, and the district court reasoned that the ditch became dangerous only when it filled with water.18 The plaintiff answered that the dangerous condition was the ditch itself, set on low ground where the possessors should have foreseen that spring rains would fill it. The dangerous condition was created at the end of the first day’s work, so enough time passed before the drowning to leave maintenance for the jury.19 The possessor’s realization is tested against what a reasonable person would foresee from the condition as it stood, so the rain that supplied the water does not restart the clock.
Within the second half of the clause, “unreasonable risk” does the screening. A foreseeable risk of harm to a child is not enough. The condition must create a risk of serious bodily harm that is unreasonable in light of the condition’s utility and the burden to the possessor of reducing or eliminating the danger.20 That makes the doctrine largely an application of ordinary negligence: if a reasonable person on the facts would see no unreasonable risk, the possessor is not negligent and is not liable.21 The screen also keeps a child’s curiosity from turning every object into a liability, for the possessor is not required to make the premises “child-proof.”22 Clause (d) supplies the explicit comparison of utility and burden against the risk, so clause (b) is best read as asking about the gravity and character of the hazard.
The risk must also be one of death or serious bodily harm, which reserves the doctrine for grave hazards.23 The floor describes the risk the possessor should have realized, and the injury the plaintiff suffered may be smaller: once clause (b) is met, the possessor is liable for any lesser injury the child in fact sustains.24 In Anderson, an excavation about ten feet deep, with water covering the bottom and metal form ties protruding from the foundation wall into the opening, was a condition a jury could find the defendant knew or should have known involved an unreasonable risk of death or bodily harm to trespassing children.25 The depth, the water, and the exposed metal each added to the gravity, and the danger lay in the thing itself.
A separate group of cases applies the unreasonable-risk requirement to a child hurt while recklessly misusing an ordinary object. Many courts deny liability where a child trespasser is injured, or injures another, while making some reckless and hazardous use of an object or condition that in itself presents no unusual danger.26 The denial rests on clause (b), quite apart from contributory negligence: the risk was not unreasonable so long as the possessor could expect the child to perceive and appreciate the consequences of the misuse.27 The danger arises from what the child chose to do with the object, and the object holds no hidden hazard for the owner to fix.
Defendants concentrate their arguments on the third step, and plaintiffs rescue cases that look too recent at the fourth. The lumber pile and the Menneti ditch mark the two ends of that range.
Child’s Inability to Appreciate Danger
Clause (c) turns from the possessor to the child. It requires that the children, because of their youth, not discover the condition or realize the risk involved in intermeddling with it or in coming within the area made dangerous by it.30 A child can fall short in two ways, by never finding the condition and by finding it without grasping what meddling with it or approaching it risks, and youth must explain either failure. The plaintiff carries the burden of proving that the child, by reason of age and inexperience, did not recognize and fully appreciate the risk.31 That allocation follows the reason for the rule. Children cannot protect themselves against dangers they do not recognize, and once a child fully understands the risk, that reason falls away.
Three questions tend to blur at this clause, and keeping them apart is most of the work of applying it. Appreciation is the question the clause asks, and it resembles assumption of risk: a child who knew of the condition, understood the danger, and went ahead anyway cannot recover.32 Contributory negligence is a different question, because it asks whether the child acted unreasonably. A child may be too young to be capable of contributory negligence, or may have acted as a reasonable child of that age would, and still lose under clause (c) because the child understood the danger completely.33 Parental supervision is the third question. In King v. Lennen, the California court observed that even very young children cannot always be kept under a parent’s watch, and that whether a parent was negligent in letting a young child play unattended near the defendant’s property will ordinarily be for the trier of fact.34 Clause (c) itself asks only what the child understood.
The court in Fields v. Henrich showed how the Restatement’s comments turn clause (c) into working tests. The duty “does not extend to those conditions the existence of which is obvious even to children and the risk of which should be fully realized by them.”35 Obviousness is judged from the child’s side, because a condition known to be dangerous to an adult will not necessarily be held dangerous to a child who lacks maturity.36 The same idea appeared under clause (b), where a risk is not unreasonable if the children whose trespass is expected can be expected to observe and fully appreciate the danger and avoid it.37 Clause (c) asks that question from the child’s side and puts the burden of proof on the plaintiff.
The comments also set a presumption for the most familiar hazards. Fire, water, and falling from a height are dangers that any child of an age to be allowed at large may reasonably be expected to understand, so the section ordinarily has no application to them unless some other factor creates a special risk that the child will not avoid the danger. Such a factor is a condition so hidden that it is not readily visible, or a distracting influence that makes it likely the child will not discover or appreciate it.38 The presumption spares possessors an impossible task, since they need not child-proof their premises against an obvious danger that every child old enough to roam already understands. Fields collects decisions applying it, including one that denied recovery for a seven-year-old’s drowning because the pond was an open and obvious danger that a child old enough to roam the fields should have comprehended.39
The exceptions explain most of the water cases that get past the presumption. Fields recognizes one where a distracting element keeps the child from seeing the inherent danger, and it recites Crawford v. Pacific Western Mobile Estates, Inc., where a six-year-old drowned in an uncovered settlement tank and the owner was liable because the effluent was dark and murky and hid its true depth, a ball floated on top, and a bridge stretched across the tank.40 Fields also recognizes an exception for a child too young to appreciate the danger, and that exception matters most with the youngest plaintiffs.41
Some courts hardened the presumption into a rule of law that any child old enough to wander appreciates certain familiar hazards. Rigid rules of that kind break down where the children likely to trespass are so young that they cannot appreciate the danger, or where the condition carries an unusual or enhanced danger that a child would not see, and the trend is to abandon them.42 King v. Lennen rejected the categorical approach in a drowning case. Liability must be decided in light of all the circumstances and not by placing cases in rigid categories by the type of condition involved. The ability to appreciate danger varies with the child’s age and mental capacity, and what matters is whether children fully understand the danger, not whether the condition is common.43 The two lines fit together once the presumption is read as tied to age, since it speaks of a child old enough to be at large. A rule tied to the label on the hazard ignores the fact the clause actually asks about, which is what this child understood.
The decided cases sort themselves by the child’s capacity. In Carmona v. Hagerman Irrigation Co. a two-year-old drowned in an irrigation canal, and the court reversed summary judgment for the irrigation company, finding genuine issues of fact under § 339 and refusing to exempt every irrigation ditch from the doctrine, although it acknowledged that it is virtually impossible to make such a ditch inaccessible to trespassing children.44 Water was involved, yet the presumption did not end the case, because a toddler is not a child at large who can be expected to understand water. A decision protecting defendants against a nine-year-old injured while jumping trains came out the other way, because the risk was obvious even to a child of that age.45 The difference lies in the plaintiffs: a two-year-old cannot be expected to recognize the danger, while a nine-year-old who jumps moving trains is taken to understand a risk that children that age understand.
Age also sets a practical ceiling. Most of the cases involve children under twelve, and a few courts have fixed an arbitrary cutoff such as fourteen. In theory the doctrine could reach a child of sixteen or seventeen when the hazard is a particularly sophisticated one, and the older the child, the fewer the dangers the child failed to understand.46 The doctrine may therefore reach a teenager and fail for a small child, or the reverse, depending on whether the particular child could understand the inherent dangers.47
Step three does the real work in water, fire, and height cases, because the presumption starts the defendant ahead and the plaintiff must bring the case within an exception. Step four matters because a child’s carelessness and a parent’s inattention are both easy to mistake for evidence on appreciation, and neither answers it.
Utility Versus Burden
Clause (d) asks whether the possessor can fairly be expected to remove the hazard that clauses (a) through (c) have established. It requires that the utility to the possessor of maintaining the condition and the burden of eliminating the danger be slight as compared with the risk to children involved.49 Two items sit on the possessor’s side of the scale: what the possessor gains by leaving the condition as it is, and what it would cost to make the condition safe. Both are set against the danger to the children likely to meet it. The earlier clauses described a grave hazard that young trespassers could not appreciate, and this clause adds the question whether guarding against that hazard is a reasonable thing to ask of an owner.
The comparison is the one ordinary negligence law already makes. Clause (d) is essentially Judge Learned Hand’s test, which sets the burden of a precaution against the likelihood and seriousness of the harm it would prevent.50 A jury in any negligence case compares the risk of injury with the utility to the defendant of keeping that risk in place, so a defendant who could remove a small risk only at great cost is not negligent.51 The clause writes that comparison into the duty because a possessor who had to make land safe for children at any price would have to give up useful conditions or child-proof whole tracts. With the comparison in place, the possessor must take a precaution only when it is cheap beside the danger.
The balance favors the child most clearly when the possessor gives up little. In Spur Feeding Co. v. Fernandez, the condition was a large rotating auger that could dismember a person and had real value to its owner, yet it could be covered at relatively small cost, so a jury could find negligence if the owner could foresee children playing nearby.52 In Banker v. McLaughlin, the condition was a dangerous pit that could easily be filled and served no useful purpose, which leaves nothing on the possessor’s side of the scale.53 The auger shows that real utility does not defeat a claim when a cheap guard preserves it, and the pit is the limiting case, with no utility and a minimal burden.
The word slight is the part of the clause most likely to mislead. Read literally, it might suggest that the plaintiff must show a wider gap between cost and risk than an ordinary negligence plaintiff, who must show only that the cost of the precaution is less than the risk. The authorities give no indication that any greater proof is required.54 Run clause (d) as the ordinary balance and do not add a hurdle the cases do not impose.
Cost also explains why courts decline to exempt a whole industry whose hazards are hard to secure. In Carmona, the irrigation company argued that it is virtually impossible to make a ditch inaccessible to trespassing children, and the court placed that concern inside clause (d). It refused to exempt any category of industry from the doctrine and treated the difficulty and expense landowners face as a factor to be weighed on the facts of each case.55 A categorical immunity would be wrong because neither side of the clause is constant across a category.
The same reasoning shapes the remote-pond cases. For farm ponds in remote areas, most courts hold that it would be an unfair burden to make owners answer to child trespassers, because fences to keep young children out may be prohibitively costly.56 The hazard is still standing water, so the difference lies on the burden side of the scale.
Clause (d) states Hand’s test and clause (e) asks whether the possessor failed to use reasonable care, so the two overlap considerably.57 The difference shows in who decides. Negligence is ordinarily a question of fact for the jury, so clause (e) usually goes to the jury, while clause (d) gives the court an additional opportunity to rule as a matter of law that the risk was not unreasonable.58 Bateman v. Mello applied the balance in that direction, reasoning that liability on its facts would carry devastating implications for landowners.59
A dissent in Franich v. Great Northern Railroad Co. shows how a plaintiff argues the balance. The dissent found clause (d) satisfied because the railroad could have kept the utility of storing its rails and still avoided the danger by posting a watchman or erecting a fence at comparatively little cost, or by placing the rails so that the accident could not happen at no extra cost.60 A dissent is no statement of the law, but it models the argument: identify a cheap alternative that leaves the possessor’s use intact and removes the hazard.
The second step is where plaintiffs win or lose most of these cases, because a cheap alternative that preserves the possessor’s use makes the burden slight, while a cost that is prohibitive at that place makes it heavy. The fourth step keeps clause (d) distinct from clause (e), since only clause (d) lets a judge end the case on the balance alone.
Failure to Use Reasonable Care
Clause (e) closes the list by requiring that the possessor fail to exercise reasonable care to eliminate the danger or otherwise to protect the children.63 The four clauses before it describe a situation: a place where child trespass is likely, a condition the possessor knows to be grave, children too young to appreciate the risk, and a precaution that would cost little beside the danger. Clause (e) is the first to ask about conduct. It names two ways to meet the duty. A possessor can eliminate the danger by filling a pit, removing a machine, or draining a tank, or can otherwise protect the children by fencing, covering, or locking the condition. A possessor who takes the steps a reasonable person would take by either route satisfies the clause, and one who takes neither does not.
The standard comes from a comment to § 339, which the court quoted in Carmona. Even when the possessor knows that children are likely to trespass, that the condition involves an unreasonable risk of harm to them, and that they are likely not to discover or appreciate the risk, there is liability only if the possessor fails to take “the steps which a reasonable [person] would take under such circumstances.”64 The limit is needed because clauses (a) through (d) describe the land and the children and say nothing about what the owner did. If those four clauses sufficed, a plaintiff who proved them would win whatever precautions the owner had taken, and the only way to avoid liability would be to keep no such condition on the land. That is strict liability, and courts have not generally adopted strict liability across the board; fault remains the usual basis of tort liability.65 The doctrine is largely an application of ordinary negligence law, in which a defendant who owes a duty of care is liable only for failing to act with ordinary care,66 and clause (e) keeps the question of what was reasonable where negligence questions ordinarily sit, with the jury.67
Clause (e) overlaps with clause (d) but asks a different question. Clause (d) weighs the utility of the condition and the burden of eliminating the danger against the risk in general, a comparison the court may resolve as a matter of law. Clause (e) asks what this possessor actually did and whether it measured up. A plaintiff who clears clause (d) has shown that a cheap precaution was available and must still show that the possessor failed to take it or an adequate substitute.
In Gregory v. Johnson the Georgia court applied clause (e) to a residential swimming pool and reversed a judgment for the pool owners. The owners had shown that the child was not invited and that they did not know of that particular child’s presence. The court held that those facts did not establish the absence of negligence as a matter of law, because issues remained about the owners’ knowledge of the presence of children who cannot appreciate the danger and about whether their failure to erect a fence, under all the circumstances, amounted to a failure of reasonable care.68 The owners’ ignorance of the individual child therefore did not settle the matter. The question was whether reasonable care called for a fence given the children the owners had reason to expect, and that question went to the jury.
The same opinion states what the duty does not demand. A landowner is not an insurer of children’s safety even when he builds a pool, and the duty is to exercise reasonable care to prevent foreseeable injury, not to guard against every injury or against remote and improbable ones.69 The court added that, once the first four conditions are found, the plaintiff must also prove clause (e), and that erecting a fence or other enclosure around a pool is generally all that reasonable care requires of the owner.70 A fence fits the duty because it answers the specific, foreseeable injury, a small child wandering in and drowning, at a cost the owner can bear. It does not promise that no child will ever get past it, and the law does not ask for that promise.
Reasonable care is judged by the precaution the possessor chose and the foreseeable risk it addressed, so a bad outcome proves nothing by itself. Gregory also shows that a possessor cannot win merely by pointing to the parents. The owners pleaded that the deceased child’s parents were contributorily negligent, and the court held that the parents’ negligence, like the owners’ own, was a question for the jury.71 Parental supervision is therefore a separate defense, and it does not answer whether the possessor used reasonable care.
The second step does most of the work, and the third explains why pool cases so often turn on whether an enclosure existed. With clause (e) the five elements are complete. The plaintiff must prove every clause, and a possessor who took the proportionate precaution defeats the claim even when the four earlier clauses are met.