This admiralty litigation was initiated by the Glenview Park District (Glenview) filing a petition in the district court seeking exoneration from or limitation of liability in connection with the drowning of Dr. George E. Melhus on September 30, 1972. The widow, as personal representative of the decedent's estate, filed in the action a claim against Glenview for wrongful death damages. Following a six-day bench trial, the district court denied Glenview's petition but on the merits entered judgment in Glenview's favor on the wrongful death claim. Both parties have appealed.

While there are some contradictions in the testamentary versions of the happenings leading to the death, and while we will advert in greater detail to certain specifics of the evidence which we deem dispositive of the legal issues raised, the following is the general factual background of this litigation as to which there appears to be no dispute.

In the fall of 1972 Glenview advertised a 4 hour, 15 mile canoe trip down the Fox River. In choosing a river to canoe Glenview checked an Illinois Department of Conservation booklet and narrowed the choice to two. The Kishwaukee was rejected in favor of the Fox, which was found to be a calm, slow moving river with very few obstructions as well as being wide. Under ordinary conditions the Fox River was shallow enough for wading. The Glenview supervisors had taken a reconnaissance run and two weeks before September 30 had taken a group trip through. They assumed that conditions would be the same as on the previous trips. They did not check with the Dayton Gauging Station or other sources for river conditions. The Fox on September 30 was at flood stage and at places had risen over its banks with the result that trees on the banks were in the water as well as the water being closer to overhanging branches.

The trip which the Melhus family took was advertised by Glenview as follows:

FEE: The fee will be $8.00 per person and will include: Transportation by bus to Sheridan, pick-up at Wedron and transportation back to Roos, use of canoe, paddles and life preservers, and supervision down the river.

Mrs. Melhus went to the Glenview office on September 16 and asked the secretary what canoeing skill was needed to participate in the trip. She was told that it was minimal and would be perfectly safe for her five and eight year old children. Children under 16 were not allowed on the trip unless accompanied by an adult. She paid the fees and was registered for the trip. Her husband was a nonswimmer, had been in a canoe several times as a youngster and had told his wife that he did not think canoeing was a very complex task.

On the day of the trip, the Glenview supervisors laid out eight canoes on the river bank and distributed various life jackets and ski-belts. Children were required to wear life preservers, but it was optional with adults. Dr. Melhus put on some sort of flotation equipment before embarking, but the evidence is not clear as to what his status was in this regard at the time of the fatal occurrence. Guy Bacci, one of the Glenview supervisors, told the group that his brother would be in the lead canoe, that he would be the last canoe, and that he would like everyone to stay relatively in the center of the river. No basic canoeing instructions were given, nor was there inquiry as to swimming ability or canoeing expertise of the voyagers.

During the early part of the trip the Melhus canoe was noticed as zigzagging considerably. After some 45 minutes on the water, the group approached an island. The Melhus canoe was in the center of the river, and as the family approached the island they saw trees in the water about 50 feet ahead of them. Dr. Melhus took over the sole paddling and attempted to steer the canoe to the left to avoid the trees. While still off the bank of the island, the canoe struck a low hanging limb which raised the canoe and spilled the family into the river. Dr. Melhus drowned in 8 to 10 feet of water.

Formal findings of fact and conclusions of law were not entered, but the final judgment order referred to the findings of fact and conclusions of law announced in open court. The judgment order recited that the canoe in question was not being operated under a bareboat charter, that the loss was not occasioned without the privity and knowledge of the Glenview Park District, and that Glenview, its agents and employees, were not guilty of a breach of any duty or guilty of any negligence which did or could have caused or contributed to the death of George Melhus.

The action of the district court is claimed to raise a number of issues for this court's consideration. We are satisfied, however, that the principal issue in these cross appeals is whether or not Glenview was guilty of actionable negligence. Before turning to that pivotal issue, we consider the threshold question of the scope of our review.

Melhus cites Barbarino v. Stanhope S. S. Co., 151 F.2d 553, 555 (2d Cir. 1945), and Pacific Tow Boat Co. v. States Marine Corp., 276 F.2d 745, 752 (9th Cir. 1960), to support her position that the determination of negligence, insofar as it requires the testing of particular facts against a fixed standard of conduct, is a jural act involving a question of law and is freely reviewable by this court.

Whatever may be the views of other courts of appeals, it appears that this court has already adopted a position regarding reviewability of a finding or conclusion of negligence. In re Rapp's Petition, 255 F.2d 628, 632 (7th Cir. 1958), ruled that this court has no greater scope of review in admiralty cases then under Rule 52(a) of the Federal Rules of Civil Procedure. In Commercial Transport Corp. v. Martin Oil Service, Inc., 374 F.2d 813, 817 (7th Cir. 1967), after observing that the trial court had found both Commercial and Martin at fault, this court stated that "(i)n reviewing the findings of the admiralty court, the 'clearly erroneous' rule governs." See also Schwerman Trucking Co. v. Gartland Steamship Co., 496 F.2d 466 (7th Cir. 1974).

We hold that review of a trial court's finding of non-negligence in an admiralty proceeding is governed by the clearly erroneous rule specified in Fed.R.Civ.P. 52(a). We consider the present case to be governed by McAllister v. United States, 348 U.S. 19, 20, 75 S. Ct. 6, 99 L. Ed. 20 (1954), which permits an appellate tribunal in an admiralty case to set aside the judgment below only when "although there is evidence to support it, the reviewing court on the entire evidence is left with the definite and firm conviction that a mistake has been committed."

This court has had little opportunity to review admiralty cases involving claims of negligent conduct resulting in death by drowning.* When one such case was presented in Rundell v. La Campagnie Generale Transatlantique, 100 F. 655 (7th Cir. 1900), this court found it unnecessary to examine in depth the concepts of proximate cause or duty in relation to a maritime tort cause of action. Relying on the analysis set forth in The Harrisburg, 119 U.S. 199, 9 S. Ct. 140, 30 L. Ed. 358 (1886), this court ruled that general maritime law allowed of no recovery in an action of tort for the death of a deceased occurring upon the high seas by reason of the negligence of the defendant. 100 F. at 659. Our Rundell decision cannot stand in light of Moragne v. States Marine Lines, 398 U.S. 375, 90 S. Ct. 1772, 26 L. Ed. 2d 339 (1970), in which the Supreme Court overruled The Harrisburg, supra, and created a uniform federal cause of action for maritime death.

Subsequently, in Sea-Land Services, Inc. v. Gaudet, 414 U.S. 573, 577, 94 S. Ct. 806, 811, 39 L. Ed. 2d 9 (1974), the Supreme Court observed that its Moragne decision was designed to extend to the dependents of maritime wrongful-death victims admiralty's " 'special solicitude for the welfare of those men who under(take) to venture upon hazardous and unpredictable sea voyages.' " Similarly, the Gaudet opinion suggested that the shape of the new maritime wrongful-death remedy be guided by the principle of maritime law that " 'certainly it better becomes the humane and liberal character of proceedings in admiralty to give than to withhold the remedy, when not required to withhold it by established and inflexible rules . . . .' " Id. at 583, 94 S.Ct. at 814.

Nonetheless, there is nothing in Gaudet's reference to the special solicitude extended to the dependents of maritime wrongful-death victims which qualifies in any way the necessary foundation of establishing actionable negligence, the basic ingredients of which are (a) negligence and (b) proximate causation, it being only necessary with regard to the second ingredient that the negligence have been a proximate cause. See 57 Am.Jur.2d Negligence § 129, at 480 (1971).

In the present case, the district court concluded that Glenview was not guilty of a breach of any duty or guilty of any negligence possessing causal significance. We recognize that there is evidence to support this finding, but we are left with the definite and firm conviction that a mistake has been made. We are persuaded that the policies underlying Moragne and Gaudet require us to accept Melhus's contention that the trial court erred in concluding that there was no Glenview negligence which was a proximate cause of the death by drowning of George E. Melhus.

The testimony at trial established that George Melhus was an inexperienced canoer and a non-swimmer. He died because he was unable to control the canoe in relatively calm current,1 was washed into an overhanging limb, and as a non-swimmer drowned in the increased depth of the Fox River, which was at flood stage on the day when Glenview conducted the supervised family canoe outing. Even if the trial court correctly found that the presence of the Melhus canoe along the bank and among the overhanging foliage or branches was a proximate cause of its overturn, its finding or conclusion that the canoe's presence there was not the consequence of any negligent conduct of Park District officials rested upon an erroneously restrictive application of the basic tort concepts of proximate cause and duty.2

Although the record establishes that the inability of Melhus to handle the canoe had a direct bearing on its presence among the overhanging branches at the edge of the Fox River, there is no basis for concluding that his canoeing incompetence, whatever its degree or extent, represented the sole and exclusive cause of the drowning death. Another salient factor was the depth of the river on the day in question. The district court judge expressed the view that, under most conditions, depth of the water is not a controlling factor in boating safety.3 He thought that the depth of the Fox River on the date of the tragedy was crucial only if it "result(ed) in a marked increase in the swiftness of current." Accordingly, he concluded that the depth of the river had "no clear causative effect resulting in a canoe being along the island bank and involved with the overhanging branches." Recognizing that it was obvious that the added depth of the waters brought the branches closer to the water level at river's edge, the district court judge still could "see no negligence of Park District officials, which was the causative factor resulting in Dr. Melhus's canoe being among these branches."

The only fair reading of the district court's summation of the case is that it thought the tragedy was caused by the inability of Dr. Melhus to handle the canoe and to keep it in the center of the river channel. Assuming arguendo that the evidence warrants such a conclusion, this court is unable to agree with the district court's proximate causation analysis. Generally speaking, an inquiry into proximate causation presupposes an affirmative finding of negligence. 65 C.J.S. Negligence § 103, at 1134 (1966).

Unfortunately, particularly for juries, some fuzziness has always been associated with the definitions and tests of proximate causation. Because the content of the concept of proximate cause must be determined by the peculiar facts and circumstances of the particular case, see 65 C.J.S., supra at § 103, none of the many various definitions of the concept can "afford a definite and invariable rule whereby a line can be drawn between those causes which the law regards as sufficiently proximate and those which are too remote to be the foundation of an action . . . ." Id.

Here, Dr. Melhus's assumption that he could handle a canoe adequately proved to be unfounded. Eschewing any purpose of relying on notions of contributory negligence or assumption of risk, which it duly recognized as not defeating recovery in admiralty, the district court nonetheless defeated recovery on the basis of proximate cause analysis. We find it unnecessary to chart the precise fashion in which the substantive notions underlying the tort concept of proximate cause flow into or eddy out from the related notions underlying the tort concept of duty of care. We agree with appellant Melhus that the applicable law of legal causation is much broader than the last act of an imperiled claimant and that, by ignoring the prior omissions of Glenview which contributed to the catastrophe, the district court applied an erroneously narrow concept of proximate cause. See also Petition of Kinsman Transit Company, 338 F.2d 708 (2d Cir. 1964), cert. denied, 380 U.S. 944, 85 S. Ct. 1026, 13 L. Ed. 2d 963 (1965).

It is well established that the existence of a duty to use due care is not destroyed by a corresponding duty on the part of another. See 65 C.J.S., supra at § 4(3), at 490. Inquiry as to the nature and extent of competing duties lies at the heart of any negligence case.

Here, the district court judge thought that his task was to "balanc(e) the obligations of the individuals for their own safety, as against the obligations of the Park District to provide for their own safety."4 In his effort to carry out the defined task, the district court viewed the dispositive issue as being what Glenview "knew or should have known concerning the dangerous conditions of the river at the time of embarkation; and whether under those circumstances they should have either cancelled the trip or warned the individuals involved that the trip was not as advertised or had some unusual circumstance which would cause them to reconsider their desire to participate."

Mrs. Melhus on appeal contends that her claim of actionable negligence should not be limited to the preembarkation period, arguing, inter alia, both on a negligence and a breach of warranty basis that Glenview undertook the duty of supervision but failed to do so in a manner resulting in the casualty. We have difficulty in saying that this claim is meritless;5 however, we think it sufficient for the disposition of this case to look at the situation as it was viewed by the district court.

It appears clear to us that Glenview, having undertaken this program for the public and having made preliminary investigation presumably in the interest of the safety of that public, could not just blandly assume that conditions along the river would remain immutable.

Had Dr. Melhus been properly and sufficiently warned of the dangers of overhanging branches near the edge of the river's bank, Glenview would be in a better position to argue that it had fulfilled its duty of due care. The record establishes no such warning, primarily because the co-supervisors of the trip were unaware that the Fox River was at flood stage on the day of the mishap. If Melhus's assumption that he could handle a canoe adequately ultimately proved to be unfounded, the same observation applies a fortiori to the assumption of Geoffrey Bacci, a co-supervisor, that the river conditions for that portion of the river the flotilla would traverse would be the same as they had been two weeks earlier.

The record contains the following colloquy:

Q . . .

Sir, did you assume that the river conditions for that portion of that river that you were to traverse would be the same that they had been when you were on the river two weeks before hand?

A Yes.

Q Sir, did you receive any orders or instructions from (higher Park District recreation officials) to personally call the Dayton Gauging Station or other weather bureau people for river condition information prior to the trip?

A No.

Q And did you personally make such a call?

A No.Q Do you know if any other Park District employee did?

A I don't know.

The volume of water measured at the Dayton Gauging Station for the canoe trip which Bacci had taken two weeks earlier was about two and a half times less than that measured for September 30, 1972. No Glenview agent inquired as to depth or volume calculations, yet the record indisputably establishes that the Fox River was at flood stage on the day of the mishap.

The duty on which actionable negligence is based can never be limitless; instead, duty is dictated and measured by the exigencies of the occasion or situation as they are, or should be, known to the actor, and it varies in each case as the facts and circumstances vary. 65 C.J.S., supra at § 4(2), at 486. Retrospective hindsight is an improper basis for ascertaining the scope of legal duties. Nonetheless, we are not persuaded that the present case really implicates that simple principle.

In many respects, the present case presents a question of nonfeasance rather than misfeasance. The Park District personnel appear to have assumed that the Illinois Department of Conservation's designation of novice canoe trails was sufficiently reliable as to warrant reliance thereon. Certainly, we cannot fault the Bacci brothers for conducting a reconnaissance run of the Fox River more that a month before the first of the two canoe outings. Nor can this court question the subsidiary fact that the Bacci brothers had found the Fox to be a calm slow moving river. As the persons most intimately connected with the actual task of leading the canoe flotilla, the Bacci brothers' preparatory work was well beyond the threshold of ordinary care. The record does not establish any basis for inferring that these two recreation specialists were even aware that the Park District's secretary had given an oral assurance that the trip would be perfectly safe for little children.

Nonetheless, the final preparations on the day of the family outing were not so complete. Assuming arguendo that Glenview's personnel could rely completely upon the trustworthiness of the Department of Conservation's designation of the lower Fox River as a novice canoe trail, those same personnel could not ignore the brochure's warning that "(s)treams change from year to year, season to season or even day to day; and information true today may not be true tomorrow." The heavy rains in the weeks preceding the late September canoe outing had, in fact, greatly increased the water depth. Outhanging branches jutting over the surface of the water necessarily became potentially dangerous obstructions to the success of a beginners' canoe trip. By ignoring the possibility that the increased depth might place dangerous jutting branches in the path of inexperienced canoeists, the Park District personnel created one link in the "chain of avoidable circumstances leading to the accident." See S.Rep.No.92-248, 92nd Cong., 1st Sess. (1971), 2 U.S.Code Cong. & Admin.News 1333, 1334 (1971).6

At the trial, Glenview's own expert witness opined that the only specific hazard on the section of the river between Sheridan and Wedron was the possibility of being washed into overhanging branches. This hazard is not generally considered a hazard of low water, especially on the Fox River, because usually the river "is shallow enough." In response to a question whether the existence of flood level condition made the use of the river by canoeists any more dangerous, Glenview's witness explained that his answer for a river like the Fox was negative. However, he thereupon qualified his flat negative by observing that:

The only if you could call it a danger, would be being washed into an overhanging branch, which is normal even on low-water streams, as well.

The expert witness further testified that at any time where there was any moving water and a slow current, like the conditions of the Fox, "you can be washed into an overhanging branch, which does not move, and the current will tend to take the canoe out from under you."

We agree with the district court that it is obvious that the added depth brings the branches closer to the water level at the river's edge. Although it is impossible to quantify the increased magnitude of the danger to canoeists engendered by flood level, that particular condition of the river does, at the very minimum, create more dangerous conditions. Where the leaders of a supervised canoe trip have a quick and inexpensive way to ascertain water levels by making a telephone call to a water gauging station and where representatives of the trip sponsor have orally assured potential participants that the trip is perfectly safe, admiralty law must recognize a duty of inquiry. We need not go so far as to characterize this routine canoe trip as a venture upon a hazardous and unpredictable sea voyage, see Gaudet, supra at 577, 94 S. Ct. 806, in order to locate the duty of inquiry. Based upon the testimony of Glenview's expert canoeist, the possibility of a canoe being washed into overhanging branches was eminently predictable.

Our holding regarding the matter of Glenview's duties and its breach thereof must be read within the framework of the particular facts presented in the case at bar. We need not reach questions relating to the general duties which the planners and leaders of canoe outings must invariably fulfill in order to preclude the possibility of being held liable for any loss or damage that may ensue. Our recognition that duty is dictated and measured by the occasion or situation and varies in each case with the facts and circumstances compels the conclusion that under some circumstances the sponsors of a canoe outing may legitimately bypass a technical inquiry regarding the water level of a designated canoe trail. Some of these trails may be so situated as to make the hazard of obstructing tree branches a virtual impossibility.

In our holding on the particular facts of this case we also are not intending to suggest any underlying Big Brother rationale to the effect that a governmental unit has because of its status a special duty of safeguarding its citizens against their own carelessness or follies. Glenview should have no greater duty of care in situations such as the present one than a private entrepreneur would have had; but it should be subject to no lesser standard.

The district court's finding that Glenview, its agents and employees, were not guilty of any breach of duty violates the underlying policies of Moragne and Gaudet. It was clearly erroneous to rule that the omissions of Glenview did not proximately cause the death of George Melhus. We are persuaded that the failure to check water conditions and to warn the participants about overhanging branches were negligent omissions which acted as proximately causative factors resulting in the Melhus canoe being among the branches. The district court should have found against Glenview on the actionable negligence question.

On remand, the district court should determine whether there was any contributory negligence on the part of Melhus which reduces recovery under the established admiralty principle that a plaintiff's negligence does not totally defeat recovery but may reduce on a proportional basis its amount.

Inasmuch as the district court did not reach or make any findings on the matter of contributory negligence, we will not attempt any analysis of that issue which will be, of necessity, dependent as a threshold matter upon the determination of the facts which under well established tort principles will reflect upon the negligence, if any, of Dr. Melhus which proximately contributed to his death.

III. Denial of Exoneration and/or Limitation of Liability

As indicated herein before, this litigation was instituted by Glenview's petition for exoneration from or limitation of liability under 46 U.S.C. §§ 181 et seq. The petition was denied by the district court, and Glenview conditionally cross appealed. We find no merit in the cross appeal.

Inasmuch as we have held that Glenview was guilty of actionable negligence, its liability for damages after the determination of the extent, if any, to which that liability might be reduced by contributory negligence could not perforce have arisen without its privity or knowledge.

Glenview briefly urges here its district court argument that Dr. and Mrs. Melhus were bareboat charterers, that they navigated the canoe and that it was their negligence which caused the occurrence. As we have already indicated, the negligence of Dr. Melhus, if there was such negligence proximately contributing to the accident, was not the sole proximate cause of the occurrence. Also it appears to us that in this canoeing trip there was about as much relationship to a maritime bareboat charter as there would be in the purchase of a ticket and participating in a ride in a boat of an amusement park Tunnel of Love.

In accordance with the above opinion, the judgment in favor of Glenview on the wrongful death claim is reversed and remanded for further proceedings in accordance with this opinion. The judgment denying the exoneration and limitation petition of Glenview is affirmed.7

AFFIRMED IN PART; REVERSED IN PART.