OPINION OF THE COURT
BECKER, Circuit Judge.
The appeal in this longshoreman’s personal injury case requires us to consider once again the contours of the “active” operations duty, as developed in the caselaw flowing from the landmark case
Scindia Steam Navigation Co. v. De Los Santos,
451 U.S. 156, 101 S.Ct. 1614, 68 L.Ed.2d 1 (1981), and to apply it to the facts of a stevedoring accident. The plaintiff is John Serbin, who, as the sun was rising on December 28, 1992, struggled to move a stuck piece of equipment — known as a “snatch block” — on the ship he was helping to unload. Unable to complete the job, Serbin attempted, with a eoworker, to set it down, but he was thrown from the crates he was standing on to the deck seven feet below, breaking his knee in the fall. Serbin sued in the District Court for the Eastern District of Pennsylvania under section 5(b) of the Longshore and Harbor Workers’ Compensation Act, 33 U.S.C. § 905(b), alleging negligence of the vessel’s crew that was attributable to the defendant ship. The district court granted summary judgment for the ship. Because there is a genuine issue of material fact as to whether the ship breached the active operations duty, we reverse and remand for further proceedings.
I. FACTS AND PROCEDURAL HISTORY
Serbin was a longshoreman employed by Independent Pier Company, a stevedore operating in the Port of Philadelphia.
Beginning at about midnight on December 28, 1992, Serbin’s longshore gang began unloading fruit from the M/V Atlantic Universal, a vessel owned by defendant Bora Corp. LTD (the “ship”). The ship’s cargo hold, where the fruit was located, is divided into hatches and decks. Each hatch, like a silo, runs vertically through the ship; each hatch is also split horizontally into five decks about seven feet high. Cargo, in this case fruit, is bundled into units approximately seven feet high so that each unit can fit into a deck. Separating the decks are movable hatch covers, like double doors, that form the floor and ceiling of the decks.
The ship’s crew opens and closes these hatch covers with a block and pulley system. Using a crane, the crew pulls a cable that runs through four snatch blocks — one in each corner — and then attaches to the hatch cover. After the longshoremen remove the unit of cargo from the highest deck, the crew uses the block and pulley system to fold open the next hatch cover, exposing the unit of cargo in the deck below. The snatch blocks, the moveable pulley part of the system, can pivot up (vertical position) and down (horizontal position) around hinges that attach them to the sides of the hatch covers. In order to open the hatch covers, the snatch blocks must be in the down (horizontal) position. After the hatch covers have been opened, the blocks must be returned to the up (vertical) position in order to allow the removal of the cargo below. The crew then ties the blocks to hold them in the up position. Each block weighs approximately one-hundred pounds. Unlike most snatch blocks, which are portable, the blocks on the M/V Atlantic Universal were fixed to the hatches and had metal projections extending from their hinges that served as stoppers. As with the hatch covers themselves, moving the snatch blocks is the responsibility of the crew.
At around 7:00 in the morning on December 28, Serbin returned from a “dinner” break to resume unloading the No. 3 hatch of the M/V Atlantic Universal, which had been loaded by another stevedore in Chile. Ser-bin, a forklift driver, was responsible for moving the fruit to the middle of the hatch, where a crane could lift the cargo out of the ship. As he descended to one of the lower “tween” decks, Serbin noticed that most of the cargo in the middle of the exposed deck had been unloaded, but that some units remained in the “wings.” He also noticed that one of the snatch blocks improperly remained in the down position, resting on top of one of the cargo units. Serbin concluded that the unit of cargo underneath the block
could not be removed while the block was in a down position, at least without damaging the top box of fruit. Serbin also believed that the block was unsafe where it was because “that’s where the hookup man would normally stand in the wing. If anything was to fall he had no place to run.” Therefore, he decided to move the block back to the up position.
Serbin decided that he should move the block himself instead of waiting for the crew to do it, he testified, for two reasons. First, “the fruit system has gotten very competitive on the East Coast. With that in [the] way we wouldn’t be able to send any fruit out and we would have had to wait for the crew to come down and move it and that would have been a waste of time, so I figured I can save time by moving it.” Second, he had moved blocks in the past without difficulty, albeit sometimes with the help of a fellow longshoreman, and saw no reason why he should have any problem in this case.
When Serbin tried to move the block, he found that he could not do so by himself. He asked another longshoreman, John McGoni-gle, who was working in the hold below, to help him raise it. McGonigle, incidentally, had notified a crew member when he too had noticed the problem. At all events, using a 4” x 4” piece of wood that was lying on the deck, McGonigle attempted to push the block from below, while Serbin, standing with one foot on top of the unit, tried to lift the block from above. They discovered that together they could still move the block only a little bit. As they attempted to set the block down, McGonigle lost control of the 4” x 4”, the block snapped back down on top of the unit, and Serbin was catapulted off the top of the unit onto the deck below. Serbin suffered a severe knee injury in the fall — a tibial plateau fracture — that has permanently disabled him from working as a longshoreman.
In addition to his and McGonigle’s testimony, Serbin offered the affidavits of two maritime experts: George Mara, a naval architect and marine surveyor; and James Muldow-ney, an experienced stevedore ship boss. These experts opined that the block had become stuck on the underlying unit when the ship’s crew failed to ensure that the block path was unobstructed before closing the hatch covers after the fruit was loaded in Chile. It was also the opinion of these experts that the crew should have discovered this condition both at the time it closed the hatch covers and the time it opened them in the Philadelphia port where Serbin worked. In his complaint, under section 5(b) of the Longshore and Harbor Workers’ Compensation Act, 33 U.S.C. § 905(b), Serbin alleged negligence on the part of the ship’s crew in failing to discover and correct the stuck block. Defendant, the ship owner, moved for summary judgment on the ground that Ser-bin could not establish a breach of any duty.
The district court granted the motion.
Serbin v. Bora Corp.,
No. 94-3030, slip op. at 1, 1995 WL 500622 (E.D.Pa. Aug.17, 1995). It first reasoned that an issue of material fact precluded deciding whether the “active operations” or the “turnover” duty of the Longshore Act governed in this ease.
Id.
at 9. Proceeding on the assumption that the active operations duty applied, the court reasoned that Serbin had failed to establish three of the four elements necessary to a prima facie ease. According to the district court, Serbin failed to show (1) that “a stuck block generally creates a hazard”; (2) that the condition of the block was not “obvious”; and (3) that the ship failed to take reasonable precautions because “[t]he vessel had established a mechanism for addressing problems with the blocks” and “[Serbin] attempted to fix the problem himself before the crew had a chance to remedy the problem.”
Id.
at 9-12. The district court concluded:
Plaintiff has not presented any evidence to establish that the obstructed condition of the block presented a hazard that either was known or should have been known to the crew. As a result, plaintiff cannot establish a breach of either the active operations duty or the turnover duty.
Id.
at 14. Serbin appeals. The standard of review for summary judgment motions is well known and hence we relegate it to the margin.
II. DUTIES UNDER THE LONGSHORE ACT
The Longshore and Harbor Workers’ Compensation Act, 33 U.S.C. § 905(b), establishes a comprehensive workers’ compensation program for longshoremen and their families.
Section 5(b), the provision of the Act relevant for our purposes, provides longshoremen a cause of action for injuries resulting from the negligence of a ship or its crew.
However, the Act neither specifies what acts constitute negligence nor describes the duties owed by shipowners to longshore workers: Instead, Congress intended that the scope of a shipowner’s liability would evolve under general common law principles.
See
H.R.Rep. No. 1441, 92d Cong., 2d Sess. (1972),
reprinted in
1972 U.S.C.C.A.N. 4698, 4704 (“Such issues can only be resolved through the application of accepted principles of tort law and the ordinary processes of litigation — just as they are in eases involving alleged negligence by land-based third parties.”).
The Supreme Court set forth the basic framework for the Act’s operation in
Scindia Steam Navigation Co. v. De Los Santos,
451 U.S. 156, 101 S.Ct. 1614, 68 L.Ed.2d 1 (1981). “This duty extends,” the Court explained, “at least to exercising ordinary care under the circumstances to have the ship and its equipment in such condition that an expert and experienced stevedore will be able by the exercise of reasonable care to carry on its cargo operation with reasonable safety to persons and property_”
Id.
at 166-67, 101 S.Ct. at 1622.
As developed in
Scindia
and subsequent cases, the Longshore Act imposes three duties on shipowners: (1) the “turnover” duty; (2) the “active operations” duty; and (3) the “intervention” duty. The primary differences between these duties turn on scope of conditions or events for which the ship is responsible. Which duty applies, in turn, depends on the timing of the cargo operation (i.e., has it begun?) and the control over the area or instrumentality in question (i.e., does the ship or the stevedore have control?). The turnover duty comprises “both a duty to provide safe conditions and a corollary duty to warn of known, nonobvious hazards” in instrumentalities and areas “turned over” to the stevedore’s control.
Kirsch v. Plovidba,
971 F.2d 1026, 1028 (3d Cir.1992). For areas and instrumentalities remaining under the ship’s control, the active operations duty includes the turnover duty, but also requires the ship, after unloading has begun, not to take negligent actions in areas under its control that threaten the longshoremen’s safety.
See Davis v. Portline Transportes Maritime Internacional,
16 F.3d 532, 537 (3d Cir.1994). Finally, the intervention duty requires the ship to take affirmative steps to rectify hazardous conditions even though it did not create the danger and even though the danger did not exist at the point of “turnover,” at least when the ship has actual knowledge and the condition is not obvious.
See Howlett v. Birkdale Shipping Co.,
512 U.S. 92, 114 S.Ct. 2057, 129 L.Ed.2d 78 (1994).
III. DISCUSSION
A. Which Duty Applies?
The two arguably relevant duties in this case are the active operations duty and the turnover duty. Serbin argues that the primary duty implicated here is the active operations duty. The ship insists that it is the turnover duty. The district court held that disputed issues of material fact precluded summary judgment on this issue. We agree. In order for the active operations duty to apply, Serbin must establish that the area in which the injury occurred, or the instrumentality which caused the injury, was under the substantial control of the vessel.
See Davis,
16 F.3d at 540. Serbin introduced evidence that the hatches and their snatch blocks (the instrumentality which allegedly caused his injury) remained at all times under the control of the crew. Indeed, the crux of the ship’s defense is that Serbin should have waited for the crew to take care of the problem.
With the exception of the “obviousness” inquiry, however, discussed
infra,
which duty controls is not important here:
if
the block presented a hazard — whether through the turnover or the active operations theory — the ship breached its duty to Serbin. Serbin has introduced evidence of the stuck block, and the ship has produced nothing that could support a conclusion that the block became stuck after Serbin’s stevedore began unloading in Philadelphia. Moreover, determining that Serbin could prevail on any theory will be enough to overcome the summary judgment against him. Therefore, like the district court, we will analyze this issue as if the active operations duty applies.
B. Did the Ship Breach Its Duty?
In order to establish a breach of the active operations duty, a plaintiff must show that the defendant “actively involve[d] itself in the cargo operations and [1] negligently injure[d] a longshoreman, or [2] [failed] to exercise due care to avoid exposing longshoremen to harm from hazards they may encounter in areas, or from equipment, under the active control of the vessel during the stevedoring operation.”
Scindia,
451 U.S. at 167, 101 S.Ct. at 1622. In
Davis,
this Court elaborated on the “due care” requirement in prong 2 of the active operation duty, the aspect relevant here. According to
Davis,
to establish a prima facie case of breach of the operations duty, a plaintiff must show:
(1) that the vessel appreciated, should have appreciated, or with the exercise of reasonable care would have appreciated, the condition; (2) that the vessel knew, or should have known, that the condition posed an unreasonable risk of harm to a longshore worker; (3) that a longshore worker foreseeably might fail to (i) either discover the condition or apprehend the gravity and probability of the harm, or (ii) protect himself or herself from the danger; and (4) that the vessel failed to take reasonable precautionary or remedial steps to prevent or eliminate the dangerous condition.
16 F.3d at 541.
1. Knowledge
The first factor Serbin must establish, and thus that we must evaluate, is whether “the vessel appreciated, should have appreciated, or with the exercise of reasonable care would have appreciated, the condition.”
Id.
The district court concluded that Serbin had satisfied this prong of his prima facie test:
In this case, the plaintiff asserts that the “condition” at issue was that the block was obstructed in such a manner that it could not be rotated. Accepting this characterization of the condition, I conclude that the plaintiff has presented sufficient evidence from which a jury could conclude that the shipowner — by way of the vessel’s crew— knew or should have known that the block was stuck. As discussed above, there is evidence that the task of moving the blocks fell to the crew. Thus, a jury could find that the crew in the normal exercise of their duties should have discovered that the block was stuck. In addition, in the hatch where the accident allegedly occurred, three of the four blocks apparently had been turned up; a jury might infer from this fact that the crew had attempted to move the fourth block but found that it was stuck. It is therefore possible to eon-
elude that the crew had actual knowledge of the stuck condition of the block.
Serbin,
slip op. at 10. We agree with the district court’s reasoning in this respect. In addition, Serbin introduced the affidavits of two experts who opined that the ship’s crew should have discovered the stuck condition of the block both when it closed the hatches in Chile and when it opened them in Philadelphia. Therefore, like the district court, we conclude that a reasonable fact-finder could determine that the ship’s crew appreciated (or should have appreciated) the condition of the block.
2. Unreasonable risk of harm
The second issue we must confront is whether Serbin presented evidence that could establish “that the vessel knew, or should have known, that the condition posed an unreasonable risk of harm to a longshore worker.”
Davis,
16 F.3d at 541. The district court reasoned as follows:
Based on the record that has been presented, ... a jury could not conclude that the vessel should have known that the stuck block presented an unreasonable risk of harm to longshore workers. Plaintiff has not come forward with any evidence that suggests that a stuck block
generally
creates a hazard and that this hazard should have been known to the crew. The only evidence I can find in the record that an obstructed block is hazardous is the fact that in this case the plaintiff was injured attempting to move it. In order to establish that the condition was hazardous, however, the plaintiff must show more than the mere fact that an accident occurred. On the record before the court, the situation encountered by the plaintiff was hazardous not because the block was stuck but because it was stuck in conjunction with the fact much of the cargo on the deck had been removed, creating a hole into which a longshore worker could fall. In order for the crew to appreciate that the obstructed block would present this risk of harm, they would have had to anticipate that a long-shore worker would attempt to move the block after removing much of the cargo. No evidence in the record would support imputing this knowledge to the crew.
Serbin,
slip op. at 10-11. We disagree with this analysis in two major respects.
First, while a plaintiff certainly cannot rely on the mere fact that an accident happened to establish the existence of a hazard, the district court’s discussion seems to suggest that a plaintiff must introduce specific evidence beyond the dangerous condition to show that that condition is generally hazardous. We disagree. This evidence might be helpful to a plaintiffs case, but it is not necessary. For instance, a plaintiff could not rely on the mere fact that he fell on a staircase to prevail in a negligence suit against the owner. But, if the plaintiff can show that the staircase was in disrepair, the jury is entitled to draw from that evidence the reasonable inference that the staircase presented a generally hazardous condition. So too in this case. Serbin need not introduce evidence about stuck blocks generally— but instead could rely on his evidence about this particular stuck block — if a fact-finder could draw the reasonable inference that the stuck block was a general hazard.
This brings us to our second point of disagreement with the district court’s analysis: whether a reasonable fact-finder could draw the inference that the condition posed an unreasonable risk of harm. Because Serbin must show that the block posed a general hazard, he is entitled to the inferences flowing from the many (i.e., general) ways a stuck block could injure someone.
See Kleinknecht v. Gettysburg College,
989 F.2d 1360, 1369 (3d Cir.1993) (“The type of foreseeability that determines a duty of care ... is not dependent on the foreseeability of a specific event.”);
Suchomajcz v. Hummel Chem. Co.,
524 F.2d 19, 28 n. 8 (3d Cir.1975) (“The concept of foreseeability means the likelihood of the occurrence of a general type of risk rather than the likelihood of the occurrence of the precise chain of events leading to the injury.”);
Restatement (Second) of Torts
§ 435(1) (“[Tjhat the actor neither foresaw nor should have foreseen the ... manner in which [the harm] occurred does not prevent him from being liable.”).
It was not necessary, therefore, that the crew anticipate that “a longshore worker
would attempt to move the block after removing much of the cargo.” A stuck block could conceivably have injured someone in any number of ways. For instance, to name just a few of the scenarios that could have occurred in Serbin’s situation alone, the block could have (1) suddenly dislodged and sent him sprawling to the ground, causing serious injury even if the ground was not seven feet below; (2) snapped back and crushed his fingers underneath; or (3) having never budged despite Serbin’s and McGonigle’s efforts, seriously injured Serbin’s back from the strain. Thus, recognizing that the stuck block presented a general hazard would not require clairvoyance on the part of the crew that a longshoreman would hurt himself in this particular way. A fact-finder could reasonably conclude that the crew should have recognized the general danger the stuck block posed.
3. Foreseeable failure of longshoreman to protect against harm
The third question is whether Serbin introduced evidence that could support a fact finding that “a longshore worker foreseeably might fail to (i) either discover the condition or apprehend the gravity and probability of the harm, or (ii) protect himself or herself from the danger.”
Davis,
16 F.3d at 541. The district court explained:
For similar reasons, a jury could not find that a longshore worker would foreseeably fail to (1) discover the condition, (2) apprehend its gravity, or (3) protect himself or herself from the danger. Once having attempted without success to move the block, a longshore worker could be expected to realize that the block was obstructed. In fact, the plaintiff acknowledges that prior to the accident, he discovered that the block was stuck. He attempted to move it by himself but, finding that he could not, he called on his co-worker McGonigle for assistance. The plaintiff has thus presented no evidence that it was foreseeable that a longshore worker would fail to discover the obstructed condition of the block. Similarly, the plaintiff has not produced evidence that it was foreseeable that a longshore worker would fail to appreciate the gravity of the block’s stuck condition. As discussed above, the evidence that has been presented suggests that the hazard associated with the stuck block existed only in conjunction with the removal of the cargo below. A longshore worker attempting to move a stuck block while standing atop a seven-foot stack of fruit boxes can be expected to appreciate the danger of falling. It is not foreseeable that such a longshore worker would fail to protect himself from this danger.
Serbin,
slip op. at 11-12. As this passage reveals, the district court’s analysis of this prong of
Davis
essentially boils down to a test of “obviousness.” We agree with this characterization.
See Davis,
16 F.3d at 543-44 (discussing this prong in terms of “obviousness”). But we disagree with the district court’s conclusion for a number of reasons.
To begin with, this Court has consistently stated that obviousness is generally a question for the jury, not often appropriate for resolution by the court on summary judgment.
See, e.g., id.
at 540;
Kirsch,
971 F.2d at 1030. We think that reasonable minds could differ on whether an “obstructed” block presented an obvious danger under these circumstances, and that it is for the fact-finder, therefore, to decide the obviousness issue in this case.
Second, we believe that the record in this case does not establish, especially at the summary judgment stage, that Serbin knew the block was “obstructed.” Serbin testified:
Q. Did you start to move the block? Because you mentioned that
sometimes
you can move the block with just one person.
A. Myself?
Q. Yes.
A. No. That’s why Johnny had to help me. I couldn’t move it myself.
Q. Did you try to move it yourself first?
A.
Yes, and I couldn’t do it.
App. 54A-55A (emphases added).
John McGonigle, who helped Serbin try to move the block, testified:
Q. [N]ormally — I’m not talking about the day of Mr. Serbin’s accident — when you lift a block like this, is it something that one person can do?
A.
No, usually two men.
Q. When you usually do it, do you use any type of equipment or did you just do it by hand?
A. Yes, we did that a few times, different jobs.
Q. Normally, would you use a four-by-four to lift the blocks?
A. Well, whatever you could find, a four-by-four, two-by-four. If there was fruit or anything in the way, you could go over and pick it up with the chisel (i.e. fork lift) forks.
App. 78A-79A (emphasis added).
Therefore, Serbin testified only that he could not move the block by himself. He did not testify that it was “obstructed” — as opposed to being too heavy. Moreover, he answered affirmatively that he could sometimes move a block himself, providing the reasonable inference that sometimes he could not move a block himself. The evidence does not, then, establish that Serbin knew that the block was “obstructed,” i.e., stuck by metal stoppers wedged into the cargo below so that it would not move even with two men and a lever. And McGonigle testified that moving a block was usually a two-man job, presumably because the blocks are heavy, and that it was not abnormal to use levers to augment the strength of the two men. McGonigle’s further testimony that other methods were available “[i]f there was no fruit or anything in the way” could possibly indicate that he had experience with this type of situation, that Serbin presumably shared this experience, and that the “fruit in the way” created an obstruction. Or it could just indicate that the fruit barred one avenue of access to the block. These questions of inference are for the fact-finder.
Granting the reasonable inferences to Ser-bin, the testimony establishes that Serbin simply thought this to be a heavy block that would require additional assistance — including both more manpower and a lever — to move. And a fact-finder could conclude that Serbin was reasonable in his assumption. If Serbin reasonably did not apprehend the stuck condition of the block, he also would have no reason to take steps to protect himself against it. A disputed issue of material fact, therefore, precludes summary judgment on this issue.
The third reason we disagree with the district court’s conclusions about foreseeable failure of the longshoreman to protect against harm is that, if the active operations duty applies, obviousness is not a complete bar to liability.
See Davis,
16 F.3d at 543-45. In
Davis,
we held that the potential obviousness of a danger — in that case a “grease and ice spot” — would not relieve a shipowner of its active operations duty to provide reasonably safe conditions under the Longshore Act, but rather could be taken into account in apportioning comparative negligence.
Id.
at 540. “[E]ven if we shared the district court’s view that a reasonable jury must conclude that the danger was obvious, known to Davis, and easily avoidable,” this Court said, “we still would not affirm its order granting [the defendant] summary judgment because we cannot conclude as a
matter of law that [the defendant] was 0% and Davis was 100% at fault.”
Id.
We concluded in
Davis
that both the structure and the legislative history of the Longshore Act demonstrated that Congress, in enacting the Act, rejected the common law tort doctrines of contributory negligence and assumption of risk in favor of the admiralty concept of comparative negligence.
Id.
at 544. A complete bar to recovery for obvious dangers, we reasoned, would be inconsistent with the Act because it would effectively implement these outmoded and congressionally rejected doctrines.
Id.
Rather, a ship could be at fault for failing to correct an unreasonable danger even if the longshoreman was also at fault for failing to avoid it: “[I]t is fundamental that there may be more than one proximate cause of an injury.”
Id.
(quoting
Moore v. M.P. Howlett, Inc.,
704 F.2d 39, 43 (2d Cir.1983)).
The ship contends that a subsequent decision of the Supreme Court,
Howlett v. Birkdale Shipping Co.,
512 U.S. 92, 114 S.Ct. 2057, 129 L.Ed.2d 78 (1994), effectively overruled
Davis.
We disagree.
Howlett
held that obviousness was a bar to liability under a different aspect of the Act — the turnover duty. Under this duty, the ship need warn only of “latent hazards in the cargo stow,” the Court said, because “[t]o impose a duty upon vessels to exercise scrutiny over a cargo loading operation to discover defects that may become hidden when the stow is complete would require vessels to inject themselves into matters beyond their ordinary province.”
Id.
at -, 114 S.Ct. at 2066. The Court made clear that the scope of the turnover duty with respect to the stow is “narrow” because “the cargo stow is separate and distinct from other aspects of the ship.”
Id.
at - - -, 114 S.Ct. at 2066-67. In contrast, “[t]he vessel’s responsibilities to inspect [the ship itself, and its gear, equipment and tools] are commensurate with its access and control.”
Id.
at -, 114 S.Ct. at 2066. “Because the vessel does not exercise the same degree of operational control over, and does not have the same access to, the cargo stow,” the Court concluded, “its duties with respect to the stow are limited by comparison.”
Id.
at -, 114 S.Ct. at 2067. Thus, the Court held that, if the hazard in that case — a sheet of clear plastic in the cargo stow — was obvious to a competent stevedore, summary judgment would be appropriate for the ship.
Moreover, the seminal case in this area,
Scindia,
also suggested that the obviousness bar to liability under the turnover duty did not apply under the active operations duty.
Compare
451 U.S. at 167, 101 S.Ct. at 1622 (turnover duty: “[I]f [the ship] fails at least to warn the stevedore of hidden danger which would have been known to him in the exercise of reasonable care, he has breached his duty-”) (emphasis added)
with id.
(active operations duty: “[T]he vessel may be liable if it actively involves itself in the cargo operations and negligently injures a longshoreman or if it fails to exercise due care to avoid exposing longshoremen to harm from hazards they may encounter_”). By omitting the modifier “hidden,” the
Scindia
Court seems to have indicated that the active operations duty is not limited to nonobvious dangers. In interpreting the Act as doing away with an “obviousness” bar to recovery under the
active operations
duty in
Davis,
we distinguished the narrower scope of the ship’s responsibility under the turnover and intervention duties. 16 F.3d at 537. “This formulation lies in stark contrast to the rule applicable when the vessel does not actively involve itself in the cargo operations,” we explained, “in which event the vessel may rely and depend on the experience and expertise of the stevedore.”
Id.
Indeed, in
Davis
itself we recognized that this Court had already decided that obviousness was a bar to liability under the turnover duty: “The focus on obviousness in
Kirsch,
971 F.2d at 1031, was linked to the turnover duty and in
Derr v. Kawasaki Kisen K.K.,
835 F.2d 490, 496 (3d Cir.1987),
cert. denied,
486 U.S. 1007, 108 S.Ct. 1733, 100 L.Ed.2d 196 (1988), to the duty to warn; here, we must come to grips with the active operations duty, a duty which contrasts materially from the duties
Derr
and Kirsch
considered.” 16
F.3d at 540 (citations omitted). We explained the different nature of the active operations duty:
When, however, the hazard occurs due to the vessel’s active operations, as is plausibly the case here, it no longer is proper for the vessel to defer to the stevedore’s expertise in handling cargo. The problem of apportioning responsibility between the vessel and stevedore by manipulating the vessel’s standard of care to account for both entities disappears, because the vessel is in such events responsible for the injury, and liability, if any, should attach to it according to its comparative fault....
In short, unlike with the turnover duty, which generally applies to hidden defects in cargo areas, the vessel cannot rely on the stevedore’s expertise to protect its workers from the vessel’s active operations.
Id.
at 548 (emphasis added).
Thus, we conclude that a fact-finder could reasonably determine that the obstructed condition was not obvious. We also hold that, if the active operations duty governs this case, obviousness will not bar liability, but rather will factor into a determination of comparative negligence.
I. Reasonable steps to avoid harm,
The final issue we must evaluate is whether the ship “failed to take reasonable precautionary or remedial steps to prevent or eliminate the dangerous condition.”
Id.
at 541. According to the district court:
[Pjlaintiff has not shown that defendant “failed to take reasonable precautionary or remedial steps to prevent or eliminate the dangerous condition.” The vessel had established a mechanism for addressing problems with the blocks: as the plaintiff acknowledges, it had made it known that the blocks were the province of the crew. Indeed, the application of the active operations duty proceeds, as discussed above, on the assumption that the blocks were the responsibility of the crew. McGonigle had. already alerted a crew member to the fact that one of the blocks was in the down position. Plaintiff, however, attempted to fix the problem himself before the crew had a chance to remedy the problem. Plaintiff cannot show that defendant breached the active operations duty, and this theory must be rejected.
Serbin,
slip op. at 12. We disagree. Ser-bin contests that the vessel had not established a “mechanism” for dealing with block problems, and we can find nothing in the record to support the ship’s and the district court’s assertion that it did. Although the active operations duty does proceed “on the assumption that the blocks were the responsibility of the crew,” Serbin introduced evidence that it was customary for longshoremen to remove hazardous conditions themselves (ineluding those involving blocks and hatch covers) so as to unload the ship quickly and efficiently. If this is true — as we must assume on summary judgment—
the ship was on notice that a competent longshoreman, perhaps unable to ascertain that the block remained “down” because it was stuck, would attempt to move it.
McGonigle’s testimony that he notified a member of the ship’s crew of the problem is also not dispositive. That another longshoreman notified a crew member of the block’s incorrect position does nothing to establish that the ship took reasonable steps to rectify it. MeGonigle testified that he had no idea if the crew member would take care of the block — or indeed whether he even spoke English — and the record contains no evidence that the ship took any steps toward taking care of the block. And even if the ship had a system in place, and had taken steps to move the block (or would have taken steps, given more time), these hypothetical factors cannot be enough to establish “reasonably precautionary or remedial measures” as long as the ship could reasonably foresee, as we have concluded that it could, that, despite such a system, a longshoreman might seek to correct the problem himself. Under these circumstances, the ship was left with two options: (1) inspecting and remedying dangerous conditions before the stevedore began cargo operations, or (2) making clear that longshoremen were not themselves to correct problems on the ship. This record permits a reasonable inference that the ship did not do either.
The ship also argues that it could reasonably rely on Serbin to stop working pending the moving of the block because OSHA regulations require him to do so. These regulations, the ship contends,
specifically state that when there is a problem with a hatch cover “that would jeopardize the safety of the [longshoreman, the problem] shall be reported at once to the officers in charge of the vessel.” 29 C.F.R. § 1918.31(e) (emphasis added). Furthermore, the OSHA regulations state that “[p]ending replacement or repairs by the vessel, work shall not be performed in the section containing the unsafe covers or in adjacent sections unless the flooring is made safe.”
Id.
The problem with this argument is that it assumes one of the ultimate issues in this litigation: Serbin’s knowledge of the condition. Because the regulations require reporting of “hatch cover” problems that would “jeopardize the safety” of longshoremen— rather than all “hatch cover” problems — Ser-bin would need to have knowledge of the block’s dangerous condition in order to comply. Serbin, of course, denies having such knowledge, and the ship introduced no evidence to the contrary. Therefore, even assuming that a regulation governing “hatch covers” also covers “snatch, blocks,” this argument is unavailing. A fact-finder could reasonably conclude that the ship failed to take reasonable steps to rectify the block’s condition.
IV. CONCLUSION
Serbin has produced evidence from which a reasonable fact-finder could conclude that the ship breached its duty to him under section 5(b) of the Longshore Act. We will, therefore, reverse the order of the district court granting summary judgment and remand for further proceedings consistent with this opinion.