OPINION

KAREN NELSON MOORE, Circuit Judge.

Plaintiffs-Appellants Ford Motor Co. (“Ford”) and its cargo insurer, Royal Insurance Co. of America (“Royal”) (collectively, “Appellants”), brought this action against Defendant-Appellee Orient Overseas Container Line Ltd. (“OOCL,” or “Appellee”), an ocean carrier, for damages arising from the loss of cargo during a transatlantic voyage. OOCL impleaded Third-Party Defendants-Appellees M/V Canmar Pride, the carrying vessel; CP Ships (UK) Ltd.; CPS No. 3 Ltd.; and CPS No. 5 Ltd. (collectively, “Third-Party Appellees”). On September 29, 2005, the district court granted partial summary judgment for OOCL and Third-Party Ap-pellees, ruling that Appellants’ claims were subject to the $500-per-package liability limitation prescribed by the Carriage of Goods by Sea Act (“COGSA”), 46 U.S.C. § 30701 et seq. Both the district court and this court authorized an interlocutory appeal of that ruling, and Appellants now argue that the district court’s ruling should be reversed. For the reasons set forth below, we REVERSE the judgment of the district court and REMAND this case for further proceedings consistent with this opinion.

I. BACKGROUND

A. Factual Background

In February 2003, Ford and OOCL entered into a “Transportation Services Main Agreement” (“TSM”) for the multimodal transport of Ford’s auto-transmission racks from Blanquefort, France to various cities in the United States. The TSM provided for transportation of the goods by land from the inland city of Blanquefort to the French port of Le Havre; by sea to Montreal, Canada; and by land to inland cities in the United States including Louisville, Kentucky; Hazelwood, Missouri; New Hope, Minnesota; and Westland, Michigan. Pursuant to the TSM, in March 2003, Ford delivered to OOCL thousands of auto transmissions held in racks within containers. OOCL loaded the cargoes on board CP Ships’ M/V Canmar Pride at the port of Le Havre. OOCL issued bills of lading for the cargoes, showing Blanque-fort as the “Place of Receipt,” Le Havre as the “Port of Loading,” Montreal as the “Port of Discharge,” and multiple inland cities in the United States as the “Place of Delivery.”

During the voyage, the vessel encountered stormy weather that washed some of the containers overboard and flooded others, damaging their contents. Appellants allege that the storm resulted in the loss of 4,387 auto transmissions and damaged 840 transmission units. Royal subsequently reimbursed Ford for the lost and damaged transmissions, pursuant to Ford’s marine-insurance policy, in the amount of $5,700,299.20.

B. Legal Background

In 1936, the United States enacted COGSA, 49 Stat. 1207 (1936) (current ver *626 sion at 46 U.S.C. §§ 30701 et seq.), which codified the 1924 International Convention for the Unification of Certain Rules of Law Relating to Bills of Lading (“the Hague Rules”). COGSA applies to “[e]very bill of lading or similar document of title which is evidence of a contract for the carriage of goods by sea to or from ports of the United States, in foreign trade.46 U.S.C. § 30701 (emphasis added). The majority of the United States’ trading partners later adopted the Hague-Visby Amendments of 1968 (“the Hague-Visby Rules”), which modified several provisions of the Hague Rules. Several of these countries also adopted a 1979 Protocol, which changed the Hague-Visby Rules from the gold standard to a limitation system based on the International Monetary Fund’s Special Drawing Rights. The United States, however, has declined to adopt either the 1968 or the 1979 amendments and has chosen to retain COGSA’s enactment of the Hague Rules. Michael F. Sturley, 2A Benedict on Admiralty § 17 (4th ed. 2004).

C. Procedural Background

On July 2, 2003, Ford and Royal filed suit against OOCL in the United States District Court for the Eastern District of Michigan, seeking to recover the value of the lost and damaged transmissions. In its Answer, OOCL asserted the $500-per-package COGSA liability limitation as an affirmative defense. OOCL subsequently filed a third-party complaint against Third-Party Appellees. Ford and Royal moved to strike OOCL’s affirmative defense, arguing that the Hague-Visby Rules, not COGSA, applied to the shipment and, therefore, that the liability limit was greater than $500 per package.

OOCL and Third-Party Appellees moved for partial summary judgment on the applicable liability limitation and on the definition of a “package” for limitation purposes, arguing that each rack of transmissions, and not each individual transmission, constituted a COGSA package. On September 29, 2005, the district court denied Appellants’ motion to strike and granted in part 1 OOCL’s and Third-Party Appellees’ motions for partial summary judgment, ruling that COGSA applied (thereby limiting liability to $500 per package) and that each rack constituted a package for the purposes of the limitation.

Ford and Royal sought, and the district court granted, certification of the liability-limitation and “package” issues for interlocutory appeal. We subsequently granted leave to appeal pursuant to 28 U.S.C. § 1292(b), and Appellants now seek reversal of the district court’s ruling.

II. ANALYSIS

This case requires us to consider, first, whether COGSA or the Hague-Visby Rules or both apply as a matter of law to the ocean voyage between Le Havre, France and Montreal, Canada. Second, having determined which law applies ex proprio vigore, 2 we must examine a complex bill of lading to determine whether Ford and OOCL contracted for the liability limits set forth in COGSA or in the Hague Visby Rules. The case presents an intellectual puzzle that we must resolve without direct precedent as guidance, and our analysis should be understood as a *627 default rule around which cargo owners and carriers can contract.

After considering relevant caselaw as well as principles of admiralty law, we reach the conclusion that the Hague-Visby Rules apply ex proprio vigore to the ocean carriage in question, between Le Havre and Montreal. We also conclude that COGSA does not apply ex proprio vigore to an ocean voyage between two foreign ports, even though the ultimate destination of a through bill of lading may be a city in the United States. But that does not end our inquiry. The Supreme Court’s decision in Norfolk Southern Railway Co. v. Kirby, 543 U.S. 14, 125 S.Ct. 385, 160 L.Ed.2d 283 (2004), affirmed the broad principle that courts should evaluate multimodal contracts in their entirety rather than treat each of the multiple stages in multimodal transportation as subject to separate legal regimes, which would present an obstacle to efficient liability rules. We therefore hold today that as a matter of federal common law COGSA liability rules apply to a multimodal maritime contract with an ultimate destination in the United States, regardless of whether the contract provides for an intermediary stop en route during the sea stage of transport or between the sea and land legs. The parties in this case, however, were free to contract for application of the liability limits set forth in either the Hague-Visby Rules or COGSA. The convoluted and contradictory nature of the contract at issue has led us to apply the doctrine of contra proferentem and to construe the bill of lading against its drafter, OOCL. We hold that OOCL and Ford contracted for application of the liability limits set forth in the Hague-Visby Rules.

A. Subject-Matter Jurisdiction

The district court had subject-matter jurisdiction over this admiralty case pursuant to 28 U.S.C. § 1333. Pursuant to 28 U.S.C. § 1292(b), this court has jurisdiction over the appeal taken from an Order of the United States District Court for the Eastern District of Michigan, entered on September 29, 2005.

B. Standard of Review

We review the district court’s order granting summary judgment de novo. DiCarlo v. Potter, 358 F.3d 408, 414 (6th Cir.2004). “Summary judgment is proper if the evidence, taken in the light most favorable to the nonmoving party, shows that there are no genuine issues of material fact and that the moving party is entitled to a judgment as a matter of law.” Macy v. Hopkins County Sch. Bd., 484 F.3d 357, 363 (6th Cir.2007); Matsushita Elec. Indus. Co. v. Zenith Radio Corp., 475 U.S. 574, 587, 106 S.Ct. 1348, 89 L.Ed.2d 538 (1986); Fed.R.Civ.P. 56(c). We must ascertain “whether the evidence presents a sufficient disagreement to require submission to a jury or whether it is so one-sided that one party must prevail as a matter of law.” Bryson v. Regis Corp., 498 F.3d 561, 569 (6th Cir.2007) (citing Anderson v. Liberty Lobby Inc., 477 U.S. 242, 256, 106 S.Ct. 2505, 91 L.Ed.2d 202 (1986)). Regardless of the evidence put forth by the nonmoving party, the moving party has the burden of establishing that “there is an absence of evidence to support the non-moving party’s case.” Patmon v. Mich. Supreme Ct., 224 F.3d 504, 508 (6th Cir. 2000) (quoting Celotex Corp. v. Catrett, 477 U.S. 317, 325, 106 S.Ct. 2548, 91 L.Ed.2d 265 (1986)). Although Appellants as well as Appellee and Third-Party Appellees in this case made cross-motions for summary judgment before the district court, Ford and Royal now appeal the grant of partial summary judgment to OOCL and Third-Party Appellees. Thus, this court must view OOCL and Third-Party Appellees as *628 the moving parties and Ford and Royal as the nonmoving parties.

C. Law Applicable to Ocean Carriage from Le Havre, France to Montreal, Canada

Numerous commentators have noted that the advent of containerization and the accompanying rise of multimodal “through” bills of lading issued by ocean carriers have complicated the legal liability regimes governing shippers’ claims. Michael E. Crowley, The Limited Scope of the Cargo Liability Regime Covering Carriage of Goods by Sea: The Multimodal Problem, 79 Tul. L. Rev. 1461 (2005). Our decision in the case before us is especially difficult because there exists little direct precedent on the issue and the lower federal court cases cited by each side, for reasons elaborated below, are neither binding on this court nor persuasive.

1. Hague-Visby Rules

The Hague-Visby Rules compulsorily apply by their own terms to the carriage of goods by sea from Le Havre to Montreal. The Visby amendments to the Hague Rules specified that they apply to bills of lading for the carriage of goods between ports in two different States (i.e., countries) if:

(a) the bill of lading is issued in a Contracting State, or

(b) the carriage is from a port in a Contracting State, or

(c) the Contract contained in or evidenced by the bill of lading provides that the rules of this Convention or legislation of any State giving effect to them are to govern the contract, whatever may be the nationality of the ship, the carrier, the shipper, the consignee, or any other interested person.

Art. 5, Protocol to Amend the International Convention for the Unification of Certain Rules of Law Relating to Bills of Lading, Feb. 23, 1968, 1977 Gr. Brit. T.S. No. 83 (Cmnd. 6944) (entered into force June 23, 1977), reprinted in 3 Thomas J. Schoenbaum, Admiralty and Maritime Law 862-63 (4th ed. 2004) [hereinafter the Hague-Visby Rules]. France is a signatory to the Hague-Visby Rules. George F. Chandler III, A Comparison of “COGSA”, the Hague/Visby Rules, and the Hamburg Rules, 15 J. Mar. L. & Com. 233, 289 (1984). Accordingly, the Hague-Visby Rules apply ex proprio vigore to the voyage because the carriage was from a port, Le Havre, in the contracting state of France.

2. COGSA

Our decision regarding whether COGSA applies by its own terms to the ocean carriage from Le Havre, France to Montreal, Canada turns on whether we read COGSA according to the plain language of the statute or according to an “ultimate destination” theory. COGSA applies to “[e]very bill of lading or similar document of title which is evidence of a contract for the carriage of goods by sea to or from ports of the United States, in foreign trade....” 46 U.S.C. § 30701 (emphasis added). “The term ‘carriage of goods’ covers the period from the time when the goods are loaded on to the time when they are discharged from the ship.” 46 U.S.C. § 30701(l)(e). “The term ‘foreign trade’ means the transportation of goods between the ports of the United States and ports of foreign countries.” 46 U.S.C. § 30701(13). Ford and Royal argue that COGSA did not apply as a matter of law to the carriage of goods between Le Havre and Montreal because neither are ports of the United States. In contrast, OOCL argues that COGSA should apply ex proprio vigore because although Mont *629 real served as the point of discharge, the ultimate destination of the cargo was cities in the United States.

Ford and Royal argue that settled admiralty law recognizes that COGSA applies by its own terms during the “tackle-to-taekle” period in the carriage of goods between foreign and U.S. ports. Indeed, we agree that to hold that COGSA applied by its own terms to the ocean voyage between Le Havre and Montreal would contradict the plain language of the statute. See, e.g., Foster Wheeler Energy Corp. v. An Ning Jiang MV, 383 F.3d 349, 355 (5th Cir.2004) (holding that “COGSA only applies compulsorily during the tackle-to-tackle period to contracts for the carriage of goods to or from U.S. ports in foreign trade”); Schramm, Inc. v. Shipco Transp., Inc., 364 F.3d 560, 565 (4th Cir.2004) (holding that “a discharge of goods occurs under COGSA when the goods are removed from the ship at the final port of destination”); Nippon Fire & Marine Ins. Co. v. M.V. Tourcoing, 167 F.3d 99, 100-101 (2d Cir.1999) (holding that COGSA applies because cargo was shipped from Japan to the United States); Pan Am. World Airways, Inc. v. Cal. Stevedore & Ballast Co., 559 F.2d 1173, 1177 n. 5 (9th Cir.1977) (observing that “COGSA has been continuously interpreted as being applicable from the time the ship’s tackle is hooked onto the cargo at the port of loading until the time when cargo is released from the tackle at the port of discharge”); Firestone Tire & Rubber Co. v. Almacenes Miramar, Inc., 452 F.Supp. 670, 671 (D.P.R.1978) (holding that “COGSA applies only to goods in shipment between the United States and a foreign port ... and only while the goods are on board ship”).

The district court opinion as well as the briefs for OOCL and Third-Party Appel-lees rely heavily on an unpublished case issued by the United States District Court for the Western District of Washington. Tarnawski v. Schenker, Inc., No. C02-5659FDB, 2003 WL 22721987, at *3 (W.D.Wash. May 6, 2003) (unpublished opinion), held that COGSA applied as a matter of law to the carriage of goods from Hamburg, Germany to Montreal, Canada as part of a shipment originating in Poland with the ultimate destination of Seattle, Washington. The Tamawski opinion provided little explanation as to why an ultimate-destination theory rather than the plain language of the statute should apply, except for citing another case decided by the United States District Court for the Central District of California. Joe Boxer Corp. v. Fritz Transp. Int’l, 33 F.Supp.2d 851, 855 (C.D.Cal.1998), held that COGSA did not apply as a matter of law to a contract for the carriage of goods from China to Guatemala via Long Beach, California. Joe Boxer, in turn, relied on People’s Insurance Co. of China v. M/V DAMODAR TANABE (In re Damodar Bulk Carriers, Ltd.), 903 F.2d 675, 680 (9th Cir.1990), which held that a distress discharge in Hawaii during a shipment of goods between ports in Chile and China was insufficient to invoke COGSA ex pro-prio vigore. On their own, these cases are not enough to persuade us to adopt an ultimate-destination theory. Joe Boxer and Damodar involved goods shipped by sea to and from foreign ports, with intermediary stops at U.S. ports. Here, goods are similarly shipped between foreign ports. But although U.S. cities were the ultimate destination of Ford’s auto transmissions, the contract provided for the transmissions to be carried there over land and not by sea. In the present case, the ocean voyage did not extend from France to a United States port with an intermediary stop in Montreal; rather, the ocean voyage ended in Montreal. There is nothing in the Damodar or Joe Boxer opinions *630 alone to persuade us to hold that a port of discharge — the last stage of a shipment by sea- — -belongs in the category of mere intermediary stops signaling no change in applicable maritime law.

The other lower court cases upon which OOCL and Third-Party Appellees rely are also unpersuasive. Hartford Fire Insurance Co. v. Orient Overseas Containers Lines (UK) Ltd., 230 F.3d 549 (2d Cir.2000), concerned a multimodal transportation contract pursuant to which cargo was trucked from Wisconsin to Chicago, moved by rail to Montreal, shipped to Belgium, and then discharged to a participating carrier who had the responsibility of trucking the cargo to the Netherlands. Thieves stole the cargo during this last stage of the transport. The Second Circuit held that the parties had not extended COGSA by contract to the transport of the cargo by land from Belgium to the Netherlands. 3 Id. at 558-59. In support of an ultimate-destination theory, OOCL points to footnote 10 of the Second Circuit opinion, stating that if the “dispute had involved the seagoing portion of the transport, COGSA would apply ex proprio vigore ....” Id. at 557 n. 10. But this footnote is merely dicta and, furthermore, may contradict other dicta in the opinion suggesting that COGSA would have applied to the ocean carriage only by contract and not ex pro-prio vigore. See id. at 557 n. 9.

OOCL also cites in support of its ultimate-destination theory Gerling-Konzern v. Expeditors International Ocean, No. 99 C 8348, 2000 WL 781434 (N.D.Ill. June 15, 2000) (unpublished opinion). The United States District Court for the Northern District of Illinois held that a carrier’s bill of lading did not extend its liability under COGSA to the periods before the goods were loaded on the vessel and after the goods were discharged, during which COGSA did not apply ex proprio vigore. Id. at *3. OOCL incorrectly states that Gerling-Konzem held that COGSA applied to the carriage of goods by sea between Genoa, Italy and Montreal, Canada. Instead, Gerling-Konzem did not definitively decide this question and merely observed “that whatever liability [the defendant] may have under COGSA as a carrier would only arise during the period of ocean transport.” Id. at *4 (emphases added).

OOCL and Third-Party Appellees lastly argue that the Supreme Court’s decision in Norfolk Southern Railway Co. v. Kirby requires us to interpret COGSA according to an ultimate-destination theory. In Kirby, the Supreme Court held that admiralty jurisdiction extends to a contract dispute when the “contract is a maritime one, and the dispute is not inherently local.” Kirby, 543 U.S. at 22-23, 125 S.Ct. 385. The Court concluded that a contract for multi-modal transportation involving both sea and land carriage retains its maritime character. “[S]o long as a bill of lading requires substantial carriage of goods by sea, its purpose is to effectuate maritime commerce — and thus it is a maritime contract.” Id. at 27, 125 S.Ct. 385. In so holding, the Supreme Court rejected a spatial interpretation of maritime contracts in favor of a conceptual one. Id. at 24, 125 S.Ct. 385.

OOCL argues that the conceptual approach to admiralty taken by the Supreme Court in Kirby suggests that COGSA applies as a matter of law whenever the ultimate destination of a shipping contract is located in the United States. Appellants correctly note, however, that Kirby *631 addressed the question of subject-matter jurisdiction rather than COGSA’s applicability as a matter of law to spatial or temporal realms beyond the plain language of the statute. Kirby's holding of admiralty jurisdiction over a dispute involving the land leg of a multimodal transportation contract does not directly resolve the question of which maritime law applies in the present case. Kirby extended only admiralty jurisdiction to the inland leg and not COGSA, itself, as a matter of substantive law. Crowley, The Limited Scope of the Cargo Liability Regime, 79 Tul. L. Rev. at 1495 (citing Kirby, 543 U.S. at 36, 125 S.Ct. 385). Thus, Kirby did not suggest that a conceptual approach to admiralty means that COGSA now applies ex proprio vigore to an ocean voyage between two foreign ports, if the ultimate destination of a through bill of lading is a city in the United States.

3. Federal Common Law

Ninety years ago, the Supreme Court concluded that “Congress has paramount power to fix and determine the maritime law which shall prevail throughout the country” and that “in the absence of some controlling statute, the general maritime law, as accepted by the Federal courts, constitutes part of our national law.” Southern Pac. Co. v. Jensen, 244 U.S. 205, 215, 37 S.Ct. 524, 61 L.Ed. 1086 (1917). Thus, in the absence of a federal statute, courts may apply federal common law. Sompo Japan Ins. Co. v. Union Pac. R.R. Co., 456 F.3d 54, 74 (2d Cir.2006) (citing E. River S.S. Corp. v. Transamerica Delaval Inc., 476 U.S. 858, 864, 106 S.Ct. 2295, 90 L.Ed.2d 865 (1986)). “What constitutes a relevant federal statute is determined solely by whether a given statute applies by its own terms to a particular set of facts.” Id. Because neither COGSA nor any other United States federal statute applies by its own terms to the ocean voyage between Le Havre and Montreal, we must apply federal common law to decide the case. Other courts have applied the Hague-Visby Rules where they apply as a matter of law and where COGSA does not apply ex proprio vigore. See An Ning Jiang, 383 F.3d at 354-55 (finding that a cargo owner and carrier had contracted to invoke COGSA liability limitations “only during periods and for claims not governed by the Hague-Visby Rules” and that Hague-Visby, and not COGSA, applied ex proprio vigore to the carriage of goods from Spain to China); Craddock Int’l Inc. v. W.K.P. Wilson & Son, Inc., 116 F.3d 1095, 1107 (5th Cir.1997) (holding that the Hague-Visby Rules and not COGSA applied as a matter of law to the transport of cargo from Venezuela to Peru); In re Damodar, 903 F.2d at 680 (holding that “a distress discharge is insufficient to invoke COGSA ex proprio vig-ore ” with respect to a contract for ocean transportation from Chile to China). None of these cases, however, involved a contract for the carriage of goods that had an originating point or ultimate destination in the United States. Accordingly, they do not persuade us to apply the Hague-Visby Rules as a matter of federal common law in the instant case.

Our primary goal is to advance the predictability and uniformity of maritime liability rules, which Congress sought to achieve in 1936 by passing COGSA and which the Supreme Court promoted in its 2004 decision in Kirby. Kirby's reasoning affirms the broader principle that courts should evaluate maritime contracts in their entirety rather than treating each of the multiple stages in multimodal transportation as subject to separate legal regimes, which would be an obstacle to uniform and efficient liability rules. The Court explained that the new technology of containerization had popularized “through” bills of *632 lading, by which cargo owners entered into a single contract for multiple stages of transportation across oceans and to inland destinations. By extending admiralty jurisdiction, the Court intended to promote cargo owners’ “efficient choice” to arrange for multimodal transport via a single bill of lading. Kirby, 543 U.S. at 25-26, 125 S.Ct. 385. Although COGSA does not apply by its own terms to the carriage of goods by sea between two foreign ports, extending COGSA’s applicability to such a voyage when the ultimate destination of a through bill of lading is in the United States will promote efficient contracting between cargo owners and shippers. When the shipment of cargo has a point of delivery in the United States, we would render contracts unpredictable if we made liability rules depend upon the location of the port of discharge, which varies according to the changing needs of the carrier. Ford admits in the instant case that the decision whether to route ships via Montreal or Norfolk, Virginia is one made by OOCL; Ford cares only that OOCL offers a competitive price and delivers the goods in time. Joint Appendix (“J.A.”) (Dunn Dep. at 65-66). It would only further confuse the already complex law governing multi-modal contract rules if a mid-voyage decision by a carrier to reroute a ship destined for Norfolk through Montreal — because of bad weather or a problem with the port service — triggered different liability limits.

Kirby suggests that we should not use the particularized geography of cargo’s transport across various oceans, railways, and highways to determine the applicable maritime law. In today’s shipping industry, goods are commonly discharged at a port outside the United States and subsequently moved by truck or rail into this country. If Montreal served as an intermediary refueling stop on a shipping route between Le Havre and New York City, that stop would not lead us to apply the Hague-Visby Rules rather than COGSA as a matter of law. See Joe Boxer, 33 F.Supp.2d at 855; In re Damodar, 903 F.2d at 680. The question now becomes whether, if the stop in Montreal marks the end of the shipping stage and the beginning of the land stage of transport, that fact should trigger the application of the Hague-Visby Rules rather than COGSA as a matter of law. To begin, we observe that making the liability rule dependent on the mode of transport subsequent to the stop in Montreal does not promote legal predictability and efficient contracting. Furthermore, Kirby’s extension of both admiralty jurisdiction and federal maritime law to the land portions of multimodal contracts encourages us to recognize that “the shore is now an artificial place to draw a line.” Kirby, 543 U.S. at 25, 125 S.Ct. 385. Kirby causes us to read Damodar and Joe Boxer in a new light. We hold today that an intermediary stop en route pursuant to a multimodal maritime contract with an ultimate destination in the United States, regardless of whether the stop is during the sea stage of transport or between the sea and land legs, should not prevent the application of COGSA liability rules as a matter of federal common law. Our decision effectuates Congress’s intent when it passed COGSA in 1936 to promote uniformity in shipping. We think that applying COGSA’s liability rules to all carriage of goods by sea, in contracts for transportation with ultimate destinations in the United States, effectuates Congress’s intent in a context that Congress could never have predicted: one in which containerized transport and “through” bills of lading prevail.

D. The Bill of Lading and Liability Limitations

1. Choice of Law

As an initial matter, we must evaluate the choice of law under which we will *633 examine the terms of the bill of lading. The TSM consists of Ford’s Global Terms and Conditions and Ford’s Supplemental Ocean Transportation Terms. J.A. at 859(TSM). Clause 10 of Ford’s Supplemental Terms, in turn, provides that OOCL’s bill of lading governs OOCL’s liability for loss or damage to Ford’s commodities. J.A. at 861(TSM). We first turn to Ford’s Global Terms and Conditions. Clause 26(a) of those terms provides that: “[a] Purchase Order shall be governed by the law of Buyer’s principal place of business without regard to conflict of laws provisions thereof, and litigation on contractual causes arising from a Purchase Order shall be brought only in that jurisdiction.” J.A. at 881(TSM). Thus, Clause 26(a) suggests that we interpret the bill of lading according to the substantive law of Michigan.

The bill of lading, too, contains a choice-of-law provision. Appellants argue that the first paragraph of Clause SO of the bill of lading applies in the instant case. That paragraph states: “the rights and obligations of all parties concerned in connection with the carriage of the Goods hereunder shall be governed by and construed in accordance with English law.” J.A. at 85 (Bill of Lading at 8). However, the second paragraph of Clause 30 states that when COGSA “shall for any reason whatsoever apply compulsorily to the carriage of the Goods hereunder then this Bill of Lading, the contract contained in and/or evidenced hereby, and the rights and obligations of all parties concerned ... shall be governed by and construed in accordance with United States law.” Id. Because we have determined that COGSA applies as a matter of federal common law, the second paragraph of Clause 30 controls, which suggests that we should evaluate the terms of the bill of lading in accordance with United States law.

Accordingly, a conflict exists between Clause 26 of Ford’s Global Terms, directing us to apply Michigan substantive law, and Clause 30 of OOCL’s bill of lading, directing us to apply federal law. Well-founded principles of contract law establish that “specific terms and exact terms are given greater weight than general language” and that “separately negotiated or added terms are given greater weight than standardized terms.... ” Restatement (Second) of Contracts § 203 (1981). Following these principles, we give greater weight to Clause 30 of the bill of lading because it is the choice-of-law clause specific to the agreement for the carriage of goods negotiated between OOCL and Ford. Clause 26, by contrast, represents general, standardized language applicable to all of Ford’s contracts for production parts and non-production goods and services. In conclusion, we apply United States law in interpreting the contract, which we held above as a matter of federal common law to be COGSA. 4

COGSA allows parties to contract for higher liability limitations than the $500 per package limitation provided by the statute. 46 U.S.C. § 30701(4), (5). We thus evaluate the bill of lading to determine whether the parties contracted for the higher liability limits provided by the Hague-Visby Rules. In so interpreting the bill of lading, we look both to the federal maritime law of contracts as well as to general principles of contract interpretation. See Maru Shipping Co. v. Burmeister & Wain Am. Corp., 528 F.Supp. 210, 214 (S.D.N.Y.1981) (holding that *634 “[w]he[n] the federal law of the sea provides ... no clear precedent ..., [the court] may look to the prevailing law of the land as applied by the states.”) To ascertain these principles of contract interpretation, we look to this and other circuits’ application of both federal-common-law principles and state law.

2. Standard for Review of the Bill of Lading

We consider as a threshold matter whether a judge or jury should resolve contract disputes. “The proper interpretation of a contract is a question of law.” JPMorgan Chase Bank, N.A. v. Winget, 510 F.3d 577, 583 (6th Cir.2007) (applying Michigan law) (quoting Archambo v. Lawyers Title Ins. Corp., 466 Mich. 402, 646 N.W.2d 170, 174 (2002)). Questions of contract interpretation, including those that form the basis for the grant of summary judgment, are subject to de novo review. Meridian Leasing, Inc. v. Associated Aviation Underwriters, Inc. 409 F.3d 342, 346 (6th Cir.2005) (applying Michigan law) (citing Campbell v. Potash Corp. of Sask, Inc., 238 F.3d 792, 797 (6th Cir.2001)). “If a contract is clear and unambiguous ... there is no issue of fact to be determined.” Lincoln Elec. Co. v. St. Paul Fire & Marine Ins. Co., 210 F.3d 672, 684 (6th Cir.2000) (quoting Inland Refuse Transfer Co. v. Browning-Ferris Indus. of Ohio, 15 Ohio St.3d 321, 474 N.E.2d 271, 272-73 (1984)). In such in stances, a court should not use extrinsic evidence to “attempt to discern the intent of the parties,” but rather should determine their intent from “the plain language of the contract.” United States v. Donovan, 348 F.3d 509, 512 (6th Cir.2003). “The question of whether the language of an agreement is ambiguous is a question of law.” Id. (citing Parrett v. Am. Ship Bldg. Co., 990 F.2d 854, 858 (6th Cir.1993)). Ambiguity exists “if the language is susceptible to two or more reasonable interpretations.” City of Wyandotte v. Consol. Rail Corp., 262 F.3d 581, 585 (6th Cir.2001) (quoting D'Avanzo v. Wise & Marsac, P.C., 223 Mich.App. 314, 565 N.W.2d 915, 918 (1997)). “In making such a determination, a district court is counseled to read the contract as a whole, and to give the contract language its ordinary and natural meaning.” Id. The above principles of contract interpretation establish that ordinarily judges interpret the language of contracts as a matter of law.

If a contract contains ambiguities, it generally becomes the task of the fact-finder to use extrinsic evidence to determine the intent of the parties. “When a contract is ambiguous, it is for the jury to determine the meaning of its terms, subject to proper instructions and based upon ‘evidence of the surrounding circumstances and the practical construction of the parties.’ ” Scott v. Anchor Motor Freight, Inc., 496 F.2d 276, 280 (6th Cir.1974) (quoting Tenn. Consol. Coal Co. v. United Mine Workers of Am., 416 F.2d 1192, 1198 (6th Cir.1969)). The jury thus uses extrinsic parol evidence to resolve conflicting terms and meanings of an ambiguous contract. Lincoln Elec. Co., 210 F.3d at 684-85 (quoting Construction Interior Sys., Inc. v. Marriott Family Rests., Inc., 984 F.2d 749, 754 (6th Cir.1993)).

Where there is no genuine issue of material fact appropriate for resolution by a fact-finder, however, a court may rely on extrinsic evidence to aid in contract interpretation. The First Circuit has reached the conclusion that “a judge may determine that the extrinsic evidence which is material is uncontested, and so appropriately enter [summary] judgment.” Nadherny v. Roseland Prop. Co., 390 F.3d 44, 49 (1st Cir.2004) (citing Adria Int’l Group, Inc. v. Ferre Dev. Inc., 241 F.3d 103, 111 *635 (1st Cir.2001)). Or a court may determine that the evidence so overwhelmingly favors one interpretation of the contract that “no reasonable person could decide the contrary.” Id. (quoting Boston Five Cents Sav. Bank v. Sec’y of Dep’t of Hous. & Urban Dev., 768 F.2d 5, 8 (1st Cir.1985)). Yet, the Fifth Circuit makes clear that when extrinsic evidence creates a genuine issue of material fact as to the parties’ intent, the interpretation of the contract becomes a matter of fact appropriately resolved by a jury. Southern Natural Gas Co. v. Pursue Energy, 781 F.2d 1079, 1081 (5th Cir.1986).

In the instant case, none of the parties have argued that the contract is ambiguous or that a factual controversy exists regarding Ford’s and OOCL’s contractual intent, which would require resolution by a fact-finder. Instead, Appellants as well as Appellee and Third-Party Appellees in this case made cross-motions for summary judgment. Both sides argue that the plain language of the contract unambiguously favors the interpretations they advance. We reverse a district court’s award of summary judgment when the contract at issue is ambiguous and the appealing party has raised a genuine issue of material fact. See, e.g., NILAC Int'l Mktg. Group v. Ameritech Servs., Inc., 362 F.3d 354, 355 (6th Cir.2004). In such instances, a jury and not a judge must consider extrinsic evidence when the contract’s language does not make clear the intent of the parties. Id. at 359. By contrast, in the instant case Appellants have raised no genuine issue of material fact regarding ambiguous terms in the contract; there is no contested evidence for a jury to resolve. We therefore conclude that the only questions of contract interpretation in this case are matters of law and that we may proceed to analyze the contract.

3. Principles of Interpretation

Courts have often treated bills of lading as contracts of adhesion that are construed against the drafter, and Appellants urge us to do the same in the instant case. Ford is a large, sophisticated corporation with decades of experience engaging in international business transactions. For this reason, we hesitate to conclude definitively that the contract for carriage of goods entered into by Ford and OOCL was a contract of adhesion. See, e.g., Union Planters Bank, N.A. v. Cont'l Cas. Co., 478 F.3d 759, 766 (6th Cir.2007) (holding that although courts frequently treat occurrence-based insurance policies as contracts of adhesion, the panel would not similarly treat the contract at issue “entered into by sophisticated parties” absent evidence that the plaintiff-appellant “was prevented from negotiating” the disputed provision). But we note that several circuit courts of appeals have treated bills of lading as contracts of adhesion, without any caveats regarding the relative size and bargaining strength of the shipper and carrier. Courts resolve these ambiguities against the drafters of the bill of lading (usually the carriers), but do not hesitate to rule in favor of carriers where there are no ambiguities in the contract. See, e.g., Gamma-10 Plastics, Inc. v. Am. President Lines, Ltd., 32 F.3d 1244, 1253 (8th Cir.1994) (noting that “[although a bill of lading is a contract of adhesion, the language in APL’s bills is clear and we can only interpret it to mean what it says”); Mori Seiki USA, Inc. v. M/V Alligator Triumph, 990 F.2d 444, 448 (9th Cir.1993) (holding that although a bill of lading is a contract of adhesion and ambiguities must be construed against the carrier, no such ambiguity existed in the bill of lading in question); SPM Corp. v. M/V Ming Moon, 965 F.2d 1297, 1302 (3d Cir.1992) (holding that a bill of lading with ambiguous liability-limitation provisions *636 should be construed against the carrier); Monica Textile Corp. v. S.S. Tana, 952 F.2d 636, 643 (2d Cir.1991) (holding that “every shipper knows ... bills of lading are contracts of adhesion ambiguities in which must be resolved against the carrier” (internal quotation marks omitted)).

We do not need to reach the conclusion that the bill of lading governing the liability provisions of the contract between Ford and OOCL was a contract of adhesion, however, because we can rely on the broader principle that ambiguities in contracts should be construed against the drafter. “If a fair reading of the entire contract of insurance leads one to understand that there is coverage under particular circumstances and another fair reading of it leads one to understand there is no coverage under the same circumstances!,] the contract is ambiguous and should be construed against its drafter....” Northland, Ins. Co. v. Stewart Title Guar. Co., 327 F.3d 448, 455 (6th Cir.2003) (quoting Raska v. Farm Bureau Mut. Ins. Co., 412 Mich. 355, 314 N.W.2d 440, 441 (1982)).

As an initial matter, we must determine who should be considered the drafter in this case, a question more complicated than one might think. Ford’s legal department drafted the boilerplate, non-negotiable contract for ocean carriage that became the TSM between Ford and OOCL. J.A. at 502-04 (Dunn Dep. at 22-24). As we noted supra, Clause 10 of Ford’s Supplemental Ocean Transportation Terms specifies that OOCL’s bill of lading governs the liability of OOCL as the carrier. J.A. at 524 (Dunn Dep. at 72). There is no evidence that OOCL made its bill of lading absolutely nonnegotiable, but the uncontested evidence is that, although Ford had the opportunity to accept or reject the bill of lading, its terms were predetermined by OOCL and not open for negotiation. J.A. at 509-10 (Dunn Dep. at 31-32). We are therefore faced with the question of whether we should consider Ford the drafter for purposes of applying the doctrine of contra proferentum, 5 or whether we should consider OOCL the drafter as the author of the specific instrument in question. Restatement (Second) of Contracts § 206 (1981), however, makes clear that in choosing among meanings of a specific term within a larger agreement, courts should generally prefer that meaning “which operates against the party who supplies the words.” Accordingly, we apply the doctrine of contra proferentum against OOCL.

Even when this and other circuits have not termed bills of lading contracts of adhesion, federal appellate courts have nonetheless resolved ambiguities in admiralty contracts against the drafters of these contracts. See, e.g., Navieros Oceanikos, S.A. v. S.T. Mobil Trader, 554 F.2d 43, 48 (2d Cir.1977); Associated Metals & Minerals Corp. v. M/V Vishva Shobha, 530 F.2d 714, 718 (6th Cir.1976) (resolving ambiguity in the bill of lading “against the party ... which prepared the instrument”); cf. Vistar, S.A. v. M/V Sea Land Express, 792 F.2d 469, 471 (5th Cir.1986) (holding that “ambiguities in the contract unresolved at trial are to be resolved against the party who drafted the contract”). Certainly, we should apply the doctrine of contra proferentem only “ ‘so long as other factors’ relevant to interpretation are ‘not decisive[.]’ ” B.F. Goodrich *637 Co. v. U.S. Fitter Corp., 245 F.3d 587, 597 (6th Cir.2001) (citing Restatement (Second) of Contracts § 206 cmt. a (1981)). The instant case, however, as we explain below, involves a decidedly ambiguous contract and no evidentiary dispute that might illuminate the parties’ intent as to whether COGSA or Hague-Visby liability limitations apply.

4. The Meaning of the Contract

Before explicating the meaning of the contract, we observe that the TSM and the bill of lading, in particular, represent to us extremely poor drafting. The terms of the bill of lading are confused, conflicting, and opaque. Perhaps, Ford and OOCL as repeat players in the shipping industry are familiar with the byzantine labyrinth of clauses meant to govern liability for multi-modal international transport. That they now ardently argue for two opposing interpretations of the contract language, however, belies the notion that they shared any common assumptions about the contract’s meaning. Regardless, the parties now rely on this court to resolve the dispute between them and we are placed at a disadvantage in so doing because of the TSM’s lack of clarity. The parties’ sloppy drafting may represent carelessness or an attempt by each side to disguise its own intentions and evade hard negotiation. In either ease, we urge Ford and OOCL to take far more care in drafting future mul-timodal shipping contracts, to avoid expensive litigation and unnecessarily consuming this court’s resources.

The district court’s grant of summary judgment to OOCL turned, in part, on its determination that Ford did not offer any extra consideration for the increase in OOCL’s liability resulting from the use of Montreal as the port of discharge. Royal Ins. Co. of Am. v. Orient Overseas Container Line Ltd., 408 F.Supp.2d 415, 422 (E.D.Mich.2005). Appellants, in contrast, argue that freight rates were higher when Montreal served as the port of discharge than when OOCL used a U.S. port. There does exist some cursory evidence that Ford and OOCL agreed upon higher rates to Montreal. A chart in the bill of lading lists consistently higher rates when the destination port is Montreal than when it is a port in the United States. J.A. at 74-75 (Rates and Charges). Without further explanation, however, this chart is not sufficient to determine that Ford and OOCL contracted for higher rates to Montreal for the purpose of covering OOCL’s increased liability. Appellants have failed “to produce ‘more than a mere scintilla of evidence’ which would create a genuine dispute for the jury.” Leary v. Daeschner, 349 F.3d 888, 897 (6th Cir.2003) (quoting Thompson v. Ashe, 250 F.3d 399, 405 (6th Cir.2001)). Neither is OOCL’s blanket assertion that Ford did not bargain for higher liability rates adequate to persuade us that such an agreement did not exist. The problem in this case is that while the contract at issue is far from clear, the parties have simply made legal arguments and have not produced any evidence as to their contractual intent. In the absence of sufficient evidence to establish a genuine issue of material fact regarding the parties’ intent, and because the parties have not averred that they would provide further evidence at trial, we apply the rule of contra proferentum in construing the ambiguous terms of this contract. See Francosteel Corp. v. M/V Pal Marinos, 885 F.Supp. 86, 89-90 (S.D.N.Y.1995).

Ford and Royal argue that Clause 4(B)(2)(a) of the bill of lading mandates that if the loss or damage occurs at a known time or place, then OOCL’s liability shall be that which is provided for by any international convention or national law that would have applied had the parties entered into a separate contract for that *638 stage of transport. In the instant case, however, Clause 4(C) rather than Clause 4(B)(2)(a) is controlling. Prior language within 4(B)(2)(a) provides for the application of Clause 4(C) “where loss or damage occurs to the Goods from the time when the Goods are loaded on board the Vessel at the Port of Loading until the time when the Goods are discharged from the Vessel at the Port of Discharge.” J.A. at 79 (Bill of Lading at 2). Because both parties agree that the loss and damage to Ford’s auto transmissions occurred at sea, somewhere between Le Havre and Montreal, Clause 4(C) governs our interpretation of OOCL’s liability.

The parties contest the meaning of Clause 4(C), the Clause Paramount. Clause 4(C) provides as follows:

All carriage under this Bill of Lading ... shall have effect subject to any legislation enacted in any country making the Hague or Hague-Visby Rules compulsorily applicable and in the absence of any such legislation in accordance with the Hague Rules or COGSA in the case of carriage to or from the United States of America.

J.A. at 80 (Bill of Lading at 3). Appellants argue that COGSA applies only as a default, when legislation making the Hague or Hague-Visby Rules compulsorily applicable is not present. OOCL argues, in contrast, that because COGSA enacted the Hague Rules, reading Clause 4(C) in the manner urged by Appellants would render the last portion superfluous. In other words, OOCL suggests that if the contract was for the carriage of goods to the United States, both the first and second portions of Clause 4(C) would make COGSA applicable. But OOCL’s argument for superfluity cuts against it as well as in its favor. In the case of carriage to the United States, the first portion of Clause 4(C) becomes redundant given the application of COGSA in the last portion of the Clause.

We agree with neither the interpretation of Clause 4(C) advanced by Ford and Royal nor that advanced by OOCL. The most reasonable interpretation of Clause 4(C) is that it anticipated three distinct situations: (1) carriage to and from two countries other than the United States subject to legislation in at least one of those countries making the Hague Rules or Hague-Visby Rules compulsorily applicable; (2) carriage to and from countries that made neither applicable, in which case the Hague Rules would apply as a default; (3) carriage between the United States and a foreign country, whereby COGSA would apply as a matter of law. This interpretation of Clause 4(C) explains the function of the language “in the absence of any such legislation” by recognizing the Clause’s intention to make the Hague Rules the default liability scheme, absent effective legislation raising liability in accordance with the Hague-Visby Amendments. Our interpretation of Clause 4(C) also explains why there exists some redundancy of language when the carriage is to the United States. Our interpretation of Clause 4(C) yields the initial conclusion that the ocean carriage at issue between Le Havre and Montreal exemplifies the first anticipated situation because both France and Canada make Hague-Visby Rules compulsorily applicable.

Nevertheless, we must consider whether the multimodal contract at issue actually exemplifies the third anticipated situation because the United States constitutes the ultimate destination of the goods, despite the ocean carriage’s end in Montreal. OOCL argues that “COGSA applies to all carriage to and from the United States of America, whether by truck, rail, etc., and not merely when the port of loading or the port of discharge is a U.S. port.” OOCL *639 Br. at 27. If we read narrowly the language in Clause 4(C) “to or from the United States,” then we would reach the conclusion that the carriage of goods by sea from Le Havre to Montreal did not involve carriage to the United States. Accordingly, the Hague-Visby Rules would apply because France is a contracting state and, under the first portion of Clause 4(C), the carriage would be subject to France’s legislation making those rules compulsorily applicable. If, however, we read the language “to or from the United States” broadly, as OOCL suggests we should, then there would arise an internal conflict in Clause 4(C). The first portion of the Clause would make the Hague-Visby Rules applicable, and the last portion of the Clause would make COGSA applicable.

Because we are obligated by the doctrine of contra proferentum to construe ambiguities in the contract against OOCL as the drafter, we reach the conclusion that Clause 4(C) makes OOCL’s liability subject to the Hague-Visby Rules and not COGSA. This conclusion is in keeping with other lower federal court cases that treat the mention of the Hague-Visby Rules in clauses paramount that are similar but not identical to the one at issue in the present case as demonstrating an agreement by the carrier to accept the Hague-Visby liability limits. See, e.g., Itel Container Corp. v. M/V Titan Scan, 139 F.3d 1450, 1454-55 (11th Cir.) (reading the general clause paramount to apply the Hague-Visby Rules in part because of the presence of a forum-selection clause applying English law that was more specific than the general boilerplate language of the clause paramount), cert. denied, 525 U.S. 962, 119 S.Ct. 405, 142 L.Ed.2d 328 (1998); Francosteel Corp., 885 F.Supp. at 89-90 (construing a bill of lading against the carrier to find that the contract incorporated the Hague-Visby Rules); Associated Metals v. M/V Star Skarven, 1995 A.M.C. 505, 515 (S.D.Fla.1994) (applying the Hague-Visby Rules in keeping with a “Visby Paragraph” of the general clause paramount to shipment from the contracting state of Finland); Ilva U.S.A., Inc. v. M/V BOTIC, Civ. A. No. 92-717, 1992 WL 296562, at *2, 1993 A.M.C. 240 (E.D.Pa. Oct. 6, 1992) (finding that the first paragraph of the general clause paramount incorporated the Hague-Visby Rules into the bill of lading and the second paragraph made them compulsorily applicable); Associated Metals & Minerals Corp. v. M/V ARKTIS Sky, No. 90 Civ. 4562(JSM), 1991 WL 51087, at *5, 1991 A.M.C. 1499 (S.D.N.Y. Apr.3, 1991) (holding that the carriage of goods from Spain, which adopted the Visby Amendments, came under the scope of the second paragraph of the general clause paramount covering actions where the Hague-Visby Rules apply), vacated and remanded on other grounds, 978 F.2d 47 (2d Cir.1992); Daval Steel Prods. v. M/V Acadia Forest, 683 F.Supp. 444, 447 (S.D.N.Y.1988) (holding that the bill of lading calls for the incorporation of the higher liability limits provided by the Hague-Visby Rules when issued in a jurisdiction where those rules apply); cf. Francosteel Corp. v. M/V DEPPE EUROPE, No. 90 Civ. 1442(RPP), 1990 WL 121683, at *3, 1990 A.M.C. 2962 (S.D.N.Y.1990) (holding that Daval Steel would be controlling but for the presence of a carrier option preserving the benefits of COG-SA for the carriers). But see Acciai Speciali Terni USA Inc. v. M/V Berane, 182 F.Supp.2d 503, 508 n. 8 (D.Md.2002) (holding that the language of the general clause paramount is insufficient to demonstrate an agreement by a carrier to accept the Hague-Visby liability limits absent further evidence of this intent).

We next evaluate how Clause 4(C), the general clause paramount, interacts with Clause 4(D), which is entitled “USA Clause *640 Paramount (if applicable).” Clause 4(D)(1) reads:

If carriage includes carriage to, from or through a port in the United States of America this Bill of Lading shall be subject to COGSA, the terms of which are incorporated herein and shall be paramount throughout carriage by sea and the entire time that the Goods are in the actual custody of the Carrier or its sub-contractor at the sea-terminal in the United States of America before loading onto the Vessel or after discharge therefrom as the case may be. (emphasis added)

J.A. at 80 (Bill of Lading at 3). OOCL argues that the affidavit of Christy Beck, a customer servieeperson employed by OOCL who is familiar with the routing of goods by rail in the United States and Canada, establishes that the goods would have traveled by rail from Montreal to the Canadian Pacific Railyard in the Port of Detroit.

Appellants argue that Clause 4(D) is “expressly subjunctive and contingent” because preceded by the qualifying phrase “if applicable.” Appellants Reply Br. at 18. Appellants also argue that the purported extension of COGSA in Clause 4(D)(1) to cover transshipment by land or sea “through” a port in the United States demonstrates OOCL’s assumption that the law governing loss at sea was that enacted by the country in which the point of discharge was located. Appellants Br. at 14-15. Lastly, Appellants argue that Beck’s loose use of the term “port” contravenes both the word’s ordinary meaning and the technical definitions of the “Port of Detroit” and the “Port of Chicago,” as explained by the Executive Director of the Detroit/Wayne County Port Authority, Curtis Hertel, and the Executive Director of the Illinois International Port District, Anthony Ianello. Appellants Reply Br. at 19 (citing Hertel Deck at 2-3 and Ianello Deck at 1-2). In sum, Appellants argue that the bill of lading provided for the goods’ transport by rail from Montreal to railyards in Detroit and Chicago located outside those cities’ ports and, therefore, the invocation of COGSA in Clause 4(D)(1) does not determine OOCL’s liability.

The conflicting evidence and argument regarding the meaning of the term “port” establishes two plausible interpretations of Clause 4(D)(1). Without evaluating the relative credibility of the Beck Affidavit and the Hertel Declaration, we conclude that Hertel’s interpretation of the meaning of the “Port of Detroit” better accords with the language of Clause 4(D)(1). Clause 4(D)(1) provides for the extension of COGSA by contract to the periods before goods are loaded on to the vessel and after they are discharged, so long as they are in the custody of OOCL or its subcontractor “at the sea-terminal in the United States of America.” The latter phrase makes clear that OOCL intended the term “port” in Clause 4(D)(1) to refer to a “sea-terminal.” Although OOCL averred that the Canadian Pacific Railyard is within the Port of Detroit, OOCL never argued that the railyard is a sea-terminal.

On the other hand, we have held above that as a matter of federal common law COGSA applies to the ocean voyage from Le Havre to Montreal despite the fact that Montreal is plainly not a U.S. port. The most reasonable interpretation of Clause 4(D)(1) is that OOCL intended to extend COGSA by contract when COGSA applied as a matter of law to the ocean voyage, to periods prior to and following “tackle-to-tackle,” so long as the goods remained within the custody of OOCL or its subcontractor. Because we have held above that COGSA applies as a matter of federal common law to the carriage of goods from Le Havre to Montreal when the ultimate *641 destination of the goods is in the United States, we hold that Clause 4(D)(1) is generally applicable. We therefore do not need to resolve the competing evidence regarding the definition of the Port of Detroit, which would remain the responsibility of a jury.

Clause 4(D)(1), however, does not specify to which particular situations it applies. If we take the language from Clause 4(D)(1) — “paramount throughout carriage by sea” — out of context, then it appears that OOCL intended to make COGSA applicable to both the sea voyage and the periods before and after discharge. When the clause is read as a whole and in harmony with the other clauses of the contract, however, it becomes clear that its purpose was to extend COGSA by contract, when COGSA was already applicable to the sea voyage, to the periods “before loading [the goods] onto the Vessel or after discharge therefrom.” Clause 4(D)(1). Indeed, the attorney for OOCL admitted at oral argument that the intent of Clause 4(D)(1) was to extend COGSA beyond the tackle-to-tackle period. Because we reached the conclusion above that Clause 4(C) made the Hague-Visby Rules applicable to the sea voyage as a contractual matter, however, there is no need to invoke Clause 4(D)(1), which addresses the periods beyond the “tackle-to-tackle” period. In the instant ease, the goods were lost at sea; therefore, this is not a case in which Clause 4(D)(1) comes into operation.

OOCL argues that Clause 4(D)(4) further extends COGSA by default to all carriage when COGSA does not otherwise apply. Clause 4(D)(4) states:

Except as provided herein in Clauses 4(D)(1) and (2), and where COGSA does not apply by operation of law, Carrier’s liability will be governed by COGSA unless its liability under some other body of law applicable to the particular stage of the transport where the loss occurred is more favorable to the Carrier (with regards to defenses and limitations), in which case that other body of law will apply.

J.A. at 80 (Bill of Lading at 3) (emphasis added). Here, we have held above that COGSA applied to the sea carriage of goods between Le Havre and Montreal as a matter of federal common law. Clause 4(D)(4) standing alone, therefore, does suggest that there exists no exception to COGSA’s applicability under the bill of lading. But Clause 4(D)(4)’s provision for COGSA liability limitations conflicts with Clause 4(C)’s provision for the Hague-Visby Rules. This ambiguity once again forces us to resort to the doctrine of contra proferentum. Accordingly, we hold that Clause 4(C), which applied the Hague-Visby Rules, controls OOCL’s liability.

One might argue that Clause 4(D)(4) trumps Clause 4(C). The contention would be that with the exception of situations anticipated by Clauses 4(D)(1) and (2) and those where COGSA does not apply as a matter of law, Clause 4(D)(4) makes COGSA’s liability limitations applicable, regardless of what Clause 4(C) suggests. The key question, however, is whether Clause 4(C), the Clause Paramount, or Clause 4(D), the USA Clause Paramount, applies. The title of Clause 4(D)(4) explicitly makes that clause conditional by including the qualifying phrase “if applicable.” Given the ambiguity as to whether Clause 4(C) or Clause 4(D)(4) applies, we believe it appropriate to place the burden on OOCL as the drafter of the bill of lading to explicitly and clearly state its own liability limits.

In addition, Himalaya Clauses in OOCL’s bill of lading make Hague-Visby’s liability limitations that are applicable to OOCL similarly applicable to Third-Party *642 Appellees. Such clauses extend liability limitations to downstream parties expected to take part in the contract. Kirby, 543 U.S. at 20 & n. 2, 125 S.Ct. 385. Clause 1 of OOCL’s bill of lading extends “the benefit of all the rights and defenses provided for” in the bill of lading to second persons or entities “including without limitation, the Vessel, her owner, operator, demise, time, slot and space charterer” who is also denominated a “carrier/bailee.” J.A. at 78 (Bill of Lading at 1). Clause 25(c) provides that “the Vessel and every subcontractor of the Carrier of any nature whatsoever ... shall have the benefit of every right, defence [sic], limitation and liberty of whatsoever nature herein contained or otherwise available to the Carrier as if such provisions were expressly for its benefit. ...” J.A. at 84 (Bill of Lading at 7). Therefore, we hold that, because the Hague-Visby Rules govern OOCL’s liability, they also govern that of Third-Party Appellees.

E. The Definition of a Package or Unit

Article 2 of the Visby Amendments replaced Article IV(5) of the Hague Rules and set forth new rules for liability limits. 3 Schoenbaum, Admiralty and Maritime Law at 861. Following the amendments, Article IV(5)(a) of the Hague-Visby Rules now states that when the shipper has not inserted the value of the goods into the bill of lading, then the carrier’s liability is limited at “the equivalent of 10,000 francs per package or unit or 30 francs per kilo of gross weight of the goods lost or damaged, whichever is higher.” Id. Article IV(5)(c) states further: “Where a container, pallet or similar article of transport is used to consolidate goods, the number of packages or units enumerated in the bill of lading as packed in such article of transport shall be deemed the number of packages or units for the purpose of [limitation of liability].... ” Id. Thus, when the Hague-Visby Rules are applicable, as in the current case, a court should limit the carrier’s liability to 10,000 francs, per package or unit listed in the bill of lading.

Appellants argue that in the “Quantity” column of the bill of lading, OOCL identified and described each of the thousands of auto transmissions that Ford shipped as a “unit.” Appellants Br. at 6, 25 (citing J.A. at 98 (OOCL Bills of Lading Ex. C. at 2)). OOCL does not dispute this contention and instead argues only that, under COGSA liability limits, each rack within the container rather than each auto transmission constitutes one COGSA package. OOCL Br. at 36-38. But we have reached the conclusion that the bill of lading binds OOCL by the Hague-Visby and not the COGSA liability limits. Accordingly, we turn to the bill of lading to assess the number of units listed by OOCL. The bill of lading, however, is sufficiently confusing to make it inadvisable for us to reach a conclusion as to the total number of units listed. A total of forty-three consecutive pages in the bill of lading include “Quantity” columns with various numbers of units listed. J.A. at 98-140 (OOCL Bills of Lading Ex. C at 2^44). An understanding of what exactly these numbers reference, where the pages duplicate and should not be counted twice, and ultimately how to calculate the total number of units listed requires a factual understanding of the practices of the shipping industry and specifically OOCL’s bill of lading. Surveying and weighing such factual evidence is the job of the fact-finder. Therefore, we remand to the district court for further proceedings consistent with this opinion. We reiterate that, as a matter of law, the total number of units so listed will constitute the number of packages or units to be used to assess the limits set by the Hague-Visby Rules to OOCL’s liability. *643 Of course, the Hague-Visby Rules only set limits to liability; actual liability will depend on other relevant factors, which have not yet been addressed by the parties or the district court.

III. CONCLUSION

Because the district court erroneously interpreted the bill of lading to apply COGSA instead of the Hague-Visby Rules, and because additional briefing and fact-finding may be required before the liability limitation may be appropriately applied, we REVERSE the district court’s judgment and REMAND this case for further proceedings consistent with this opinion.

Notes

. OOCL's and Third-Party Appellees’ motions also sought summary judgment regarding whether all of the transmissions alleged to have been damaged actually were damaged, but the district court declined to decide that issue.

. The term ex proprio vigore means "[b]y their or its own force.” Black’s Law Dictionary 612 (8th ed. 2004).

. The contract in Hartford provided for the application of COGSA only in the absence of a different law directly applicable to that stage of the transport and, in fact, another law applied: the Convention on the Contract for the International Carriage of Goods by Road.

. We note that as a practical matter it makes no difference to the case whether we apply Michigan’s rules of contract interpretation or the rules established by federal common law because they are substantially similar as applied to the issues raised by this case.

. The contra proferentum doctrine applies “where contractual language is found to have more than one interpretation. If the language is subject to more than one interpretation, ambiguities are construed against the drafter of the language.” Marquette Gen. Hosp. v. Goodman Forest Indus., 315 F.3d 629, 632 n. 1 (6th Cir.2003) (applying Michigan law).