SELYA, Circuit Judge.
This case poses a puzzling question about when an affiliate is not an affiliate.
Cfi
William Shakespeare, Romeo and Juliet, act II, sc. ii (1595) (“What’s in a name? [TJhat which we call a rose [b]y any other name would smell as sweet[.]”). The district court agreed with Blinds to Go, Inc. (BTG) and its shareholders that Harvard Private Capital Holdings, Inc. (Holdings) violated their right of first refusal when it
transferred all of BTG’s preferred shares to the putative affiliate, Charlesbank Equity Fund II, Limited Partnership (the Fund). Accordingly, the court rescinded the transaction.
The district court’s decision pleased no one. Holdings and the Fund argue that they are in fact affiliates and assail the district court’s finding that the transfer
inter sese
violated the right of first refusal. For their part, BTG and its shareholders excoriate the district court’s choice of remedy. Reexamining the matter afresh, we conclude, as did the lower court, that a breach of the right of first refusal occurred. We therefore reject the appeal brought by Holdings and the Fund. We also conclude that the district court’s choice of remedy for that breach (voiding the transfer rather than decreeing specific performance) was consistent with the contract and with equitable remedial principles. We therefore reject the appeal taken by BTG and its shareholders.
I. BACKGROUND
BTG is a closely held Canadian corporation that manufactures, sells, and installs custom-made window treatments. Its seven shareholders include six Canadian corporations and Nkere Udofia, BTG’s vice-chairman.
Holdings is a not-for-profit Massachusetts corporation. Its sole member is the designee of the President and Fellows of Harvard College (Harvard). The Fund is a limited partnership organized under Massachusetts law. Its general partner is Charlesbank Equity Fund II GP, Limited Partnership (the General Partner); its limited partners are three charitable corporations wholly owned by Harvard, namely, Holdings, Phemus Corp., and Shipping Venture Corp. Structurally, the General Partner is itself a Massachusetts limited partnership; its general partner is Char-lesbank Capital Partners, LLC (the LLC), a ■ Massachusetts limited liability company owned by its individual members. The General Partner has one Class C limited partner, namely, Harvard Private Capital Properties, Inc. (Harprop), a Delaware corporation wholly owned by Harvard.
A venture capital transaction set in motion the events leading to this litigation. In 1995, pursuant to the BTG Preferred Share Purchase Agreement (the Purchase Agreement), Holdings injected $15,000,000 in capital into BTG in exchange for approximately 20,000,000 shares of BTG’s preferred stock. On December 31, 1997, the parties executed an amended and restated shareholders’ agreement (the Shareholders’ Agreement) which, along with the Purchase Agreement, governs their relationship. Among other things, the Shareholders’ Agreement provides the BTG shareholders with a right of first refusal vis-a-vis the stock owned by Holdings. The right of first refusal attaches to any transaction other than one involving an affiliate.
In or around 1998, Harvard began to restructure its investment portfolio for purposes of tax advantage and business convenience. In 2001, as part of this restructuring, Holdings’ in-house counsel, without troubling to read the relevant document, informed BTG that Holdings planned to make a permitted transfer of its BTG shares to an affiliate. Holdings proceeded to convey those shares to the Fund. The parties recorded the transfer at book value (i.e., $15,000,000). In exchange, Holdings received a 12.4% ownership interest in the Fund. Because it transferred other assets as well, Holdings’ total ownership interest in the Fund reached 52.9%.
On January 14, 2002, Holdings and the Fund sought to exercise a “put” right contained in the Purchase Agreement. That right allowed Holdings or its lawful successor in interest to demand, at either of two specified times, that BTG redeem all of the preferred shares. Under the Purchase Agreement, the redemption price was to be established through a formula emphasizing BTG’s earnings before interest, taxes, depreciation, and amortization (EBIT-DA) for the preceding twelve months.
Storm clouds began to gather when the redemption price, as tentatively calculated by BTG, proved to be far less munificent than Holdings and the Fund expected.
See Charlesbank Equity Fund II v. Blinds to Go, Inc.,
370 F.3d 151, 154-55 (1st Cir. 2004) (explicating more completely the factual background of the put and the attempted redemption). The storm broke when the Fund, invoking diversity jurisdiction,
see
28 U.S.C. § 1332(a), filed suit against BTG in the United States District Court for the District of Massachusetts. The Fund asserted common law claims arising out of an alleged manipulation of BTG’s finances with a view toward reducing the value of the put. Holdings soon joined the fray as an additional plaintiff. BTG denied the essential allegations of the complaint and posited, as an affirmative defense, that it owed nothing on the put because Holdings had breached the Shareholders’ Agreement when it transferred the shares to the Fund without honoring the right of first refusal.
On July 23, 2003, the BTG shareholders filed a separate action in the district court seeking (i) a declaration as to whether the transfer between Holdings and the Fund was a transfer to an affiliate as that term is defined in the Shareholders’ Agreement and (ii) relief for Holdings’ purported breach of the Shareholders’ Agreement. On October 15, 2003, the district court consolidated that action with the original action.
After much procedural maneuvering,
see, e.g., Charlesbank,
370 F.3d at 153, BTG and its shareholders moved for summary judgment on all claims and counterclaims. Not to be outdone, Holdings and the Fund cross-filed for partial summary judgment on the right of first refusal claim. Following a hearing, the district court, in a bench decision, granted summary judgment in favor of the BTG shareholders on the right of first refusal claim, denied the cross-motion for partial summary judgment, and reserved decision on
the remaining issues in the case. The court concluded (i) that the Fund was not an affiliate of Holdings within the contemplation of the Shareholders’ Agreement; (ii) that compliance with the right of first refusal constituted a condition precedent to the proposed transfer; (iii) that because Holdings did not abide by the right of first refusal provision, the transfer was void
ab initio;
and (iv) that the appropriate remedy was to unravel the transaction and require the Fund to return the stock to Holdings. The district court later entered a partial final judgment to this effect.
See
Fed.R.Civ.P. 54(b). These timely appeals ensued.
II. ANALYSIS
Given the district court’s detailed findings, we do not doubt our appellate jurisdiction over these interlocutory appeals.
See Spiegel v. Trs. of Tufts Coll.,
843 F.2d 38, 42-44 (1st Cir.1988) (delineating the requirements for invocation of Rule 54(b)). We therefore proceed to assess the district court’s conclusions. We divide our discussion into two segments.
A.
Operation of the Right of First Refusal.
Holdings and the Fund contest the district court’s construction of the right of first refusal provision. They contend that they are in fact affiliates as that term is defined in the Shareholders’ Agreement and that, therefore, the transaction between them never triggered, much less violated, the right of first refusal. This contention is the logical starting point for our analysis; after all, if Holdings and the Fund are correct, then the share transfer was valid, the right of first refusal was not implicated, and there would be no need for us to address the remedial aspect of the district court’s decision.
In approaching this question, we replay a familiar standard of review. “A district court may enter summary judgment upon a showing ‘that there is no genuine issue as to any material fact and that the moving party is entitled to a judgment as a matter of law.’ ”
Houlton Citizens’ Coal. v. Town of Houlton,
175 F.3d 178, 183 (1st Cir.1999) (quoting Fed. R.Civ.P., 56(c)). We review an entry of summary judgment de novo and, therefore, apply the same analytic framework here.
See id.
at 184. That framework is not affected by the existence of a cross-motion for summary judgment.
See Blackie v. Maine,
75 F.3d 716, 721 (1st Cir.1996).
On this issue, we are faced with a question of contract interpretation: do Holdings and the Fund qualify as affiliates under the Shareholders’ Agreement? Section 2.4 of that agreement creates the benchmark. It provides:
An “Affiliate” of a person or entity shall mean another person or entity that is directly or indirectly controlling, controlled by or under common control with such person or entity. “Control” shall mean the right to cast, directly or indirectly, more than 50% of the voting interests in a person or entity.
The Shareholders’ Agreement recites that it is governed by Massachusetts law and, thus, we look there for the substantive rules of decision.
Under Massachusetts law, an unambiguous contract must be interpreted according to its terms.
See Freelander v. G. & K. Realty Corp.,
357 Mass. 512, 258 N.E.2d 786, 788 (1970);
see also Fairfield 274-278 Clarendon Trust v. Dwek,
970 F.2d 990, 993 (1st Cir.1992) (applying Massachusetts law). In such a situation, contract construction presents an unadulterated question of law.
See Daley v. J.F.
White Contracting Co.,
347 Mass. 285, 197 N.E.2d 699, 702 (1964).
A contract is ambiguous only when its terms “are inconsistent on their face” or when “the phraseology can support reasonable difference of opinion as to the meaning of the words employed.”
Suffolk Constr. Co. v. Lanco Scaffolding Co.,
47 Mass.App.Ct. 726, 716 N.E.2d 130, 133 (1999) (quoting
Fashion House, Inc. v. Kmart Corp.,
892 F.2d 1076, 1083 (1st Cir. 1989)). There is no ambiguity simply because a dispute exists between the contracting parties, each lobbying for its own preferred interpretation.
Id.
We discern no ambiguity in the relevant text of section 2.4. That section defines the term “[affiliate” with considerable precision and, in doing so, not only endows the word “control” with decretory significance but also assigns that word a specific meaning. We therefore rely on the plain language of the provision to resolve the legitimacy of the transfer.
Holdings is fully controlled (under any definition of the word) by Harvard. In determining who “controls” the Fund, however, we cannot rely on general usage but, rather, must apply the specific definition agreed upon by BTG and Holdings.
See Rogaris v. Albert,
431 Mass. 833, 730 N.E.2d 869, 871 (2000) (stating that a contract’s terms will not “be taken in their plain and ordinary sense” if “otherwise indicated by the contract”);
see also Charles I. Hosmer, Inc. v. Commonwealth,
302 Mass. 495, 19 N.E.2d 800, 804 (1939) (explaining that “every phrase and clause must be presumed to have been designedly employed”). Because that definition turns on voting rights, we conclude, without serious question, that the Fund is directly controlled by the General Partnfer (after all, pursuant to the Fund’s organic document — its limited partnership agreement — only the General Partner has the power to “vote, give assent and otherwise ... exercise all rights, powers, privileges and other incidents of ownership or possession with respect to” the Fund’s assets). While Holdings is a limited partner, the limited partnership agreement specifically declares that it “shall take no part in the conduct or control of the Partnership business.”
The General Partner itself is, of course, a limited partnership. This is scant solace to Holdings and the Fund, however, as its limited partnership agreement states unequivocally that “[t]he Partnership shall be managed exclusively” by its general partner (the LLC). Thus, the LLC controls the General Partner. In turn, the LLC is controlled by a manager and various individual members (a group that excludes both Harvard and Holdings).
Given this hierarchy, it is readily evident that neither Harvard nor Holdings occupies a place in the Fund’s chain of voting control. Neither Harvard nor Holdings controls the Fund “directly” (a status reserved to the General Partner). By the same token, neither Harvard nor Holdings controls the Fund “indirectly” (a status reserved to the LLC and its members). Holdings and the Fund are, therefore, not affiliates within the narrow compass of the definition contained in the Shareholders’ Agreement.
In an effort to blunt the force of this reasoning, Holdings and the Fund make three points. First, they argue that because Harvard (through Holdings, Phe-mus, and Shipping Venture) continues to own virtually all of the beneficial interest in the Fund — 99.92%—it, in effect, “controls” the Fund. Second, they posit that because Holdings itself owns a majority interest in the Fund — 52.9%—it has the power under the partnership agreement to require the Fund to reconvey its assets to
the limited partners; this power of recon-veyance, they say, amounts to “control.” Third, Harvard, through Harprop (the only Class C limited partner of the General Partner), can direct the LLC (the general partner of the General Partner) to distribute the Fund’s assets to its limited partners, so it “controls” the Fund in that sense as well.
To be sure, these scenarios suggest “control” in a lay sense. This case, however, is not concerned with the ordinary meaning of “control.” Where the parties to a contract take pains to define a key term specially, their dealings under the contract are governed by that definition.
See J.A. Sullivan Corp. v. Commonwealth,
397 Mass. 789, 494 N.E.2d 374, 378 (1986) (recognizing that “[a] contract is to be construed to give reasonable effect to each of its provisions”). So it is here. The parties to the Shareholders’ Agreement crafted a specific definition of the word “control.” Holdings and the Fund cannot now gloss over that definition—nor can we.
See Charles I. Hosmer, Inc.,
19 N.E.2d at 804 (explaining that all phraseology in a contract “must be given meaning and effect, whenever practicable”). The mere fact that the transaction, viewed without regard to the Shareholders’ Agreement, represented a transfer from one Harvard pocket to another is not enough to override the explicit language that the parties chose to insert into the instrument.
Holdings and the Fund insist that the word “indirect,” used twice in section 2.4, expands voting control to include practical control of any kind. That word, however, cannot carry the weight that Holdings and the Fund place upon it. In context, the word refers only to corporate structure (e.g., the LLC “indirectly controls” the Fund because it “directly controls” the General Partner, which, in turn, “directly controls” the Fund). It would be unreasonable to read the word as a fundamental alteration of the precise definition of “control” that accompanies it. At bottom, then, the arguments mounted by Holdings and the Fund represent thinly veiled attempts to redefine “control” to comport with economic realities rather than with voting rights. Whatever the merits of this perspective in the abstract, we cannot countenance so blatant an attempt to rewrite a clearly defined contract term.
See Freelander,
258 N.E.2d at 788 (observing that when “[t]here is nothing ambiguous in [a contract’s] language ... [a] court cannot subvert its plain meaning”). Consequently, we conclude, as did the court below, that Holdings and the Fund are not affiliates as that term is defined in the Shareholders’ Agreement.
That ends this aspect of the matter. Since it is undisputed that Holdings did not afford the BTG shareholders the specified opportunity to exercise their fight of first refusal, the transfer violated the Shareholders’ Agreement.
B.
Choice of Remedy.
The focus of the appeal taken by BTG and its shareholders is the district court’s choice of remedy. Citing
Town of Sudbury v. Scott,
439 Mass. 288, 787 N.E.2d 536 (2003), they insist that specific enforcement of a disregarded right of first
refusal is the exclusive remedy permitted under Massachusetts law. Accordingly, their thesis runs, the district court should have ordered Holdings to offer the preferred shares to the BTG shareholders on the same terms as were made available to the Fund. We think that this thesis takes too crabbed a view of a trial court’s equitable powers under Massachusetts law.
To facilitate this phase of our inquiry, we assume for argument’s sake that the Fund’s acquisition of the shares met the initial requirements set forth in section 3.1 of the Shareholders’ Agreement (i.e., that there was an offer, which was bona fide and led to an arm’s length transaction).
See supra
note 2. This set of assumptions primes the pump and brings us directly to the question of remediation.
The anodyne that the BTG shareholders seek (specific performance) and the anodyne that the district court decreed (rescission) are both equitable remedies.
See, e.g., Kenda Corp. v. Pot O’Gold Money Leagues, Inc.,
329 F.3d 216, 224 (1st Cir.2003);
Pritzker v. Yari,
42 F.3d 53, 72 (1st Cir.1994). Two abiding truths about equitable remedies are pertinent here. First, in choosing among equitable remedies, a nisi prius court has the ability — indeed, the duty — to weigh all the relevant facts and circumstances and to craft appropriate relief on a case-by-case basis.
See Rosario-Torres v. HernandezColon,
889 F.2d 314, 321 (1st Cir.1989) (en banc). Second, in shaping an equitable remedy, a nisi prius court typically has a range of appropriate options. As long as the court’s ultimate choice falls within this range, it will withstand review even if it is not, in the appellate court’s opinion, the best option within the range.
See id.
at 324 (describing a district court’s choice of equitable remedies as “quintessentially a judgment call” and noting that, absent clear error, it does not matter whether the court of appeals might have made some other choice).
Viewed against this backdrop, it should come as no surprise that the standard of review applicable to a district court’s choice among available equitable remedies is deferential. A deferential standard is particularly appropriate where, as here, the trial court must balance conflicting factors and deal with issues of judgment.
See Charlesbank,
370 F.3d at 158. Therefore, we review the trial court’s choice among available equitable remedies for abuse of discretion.
See Texaco P.R., Inc. v. Dep’t of Consumer Affairs,
60 F.3d 867, 875 (1st Cir.1995);
Rosario-Torres,
889 F.2d at 323. Because the district court “is in a considerably better position to bring the scales into balance than an appellate tribunal,” we will not normally find an abuse of discretion unless, upon whole-record review, we are convinced that the district court committed a significant error in judgment.
Rosano-Torres,
889 F.2d at 323.
Of course, a material error of law constitutes an abuse of discretion.
Rosario-Urdaz v. Rivera-Hernandez,
350 F.3d 219, 221 (1st Cir.2003). Invoking this principle, BTG and its shareholders cite
Town of Sudbury
for the proposition that Massachusetts law limits the range of remedies available in this instance to one: specific performance. But the Massachusetts Supreme Judicial Court (the SJC) has emphasized that “[e]quitable remedies are flexible tools,”
Demoulas v. Demoulas,
428 Mass. 555, 703 N.E.2d 1149, 1169 (1998), and we do not believe that it has constrained the inherently flexible nature of equity as severely as BTG suggests.
In
Town of Sudbury,
the SJC addressed a municipality’s claim against a purchaser of agrarian land. The purchaser had paid a reduced price to the original owner on
the understanding that he would maintain the land for agricultural use.
See Town of Sudbury,
787 N.E.2d at 538. Had he bought it for any other use, the town would have been entitled to a statutory right of first refusal.
See id.
at 541.
Only months after the acquisition, the purchaser began to prepare the site for non-agricultural use and soon entered into a purehase-and-sale agreement with a third party who intended to carry out the non-agricultural use.
See id.
at 538-^40. When the town learned of the impending sale, it sought to exercise its right of first refusal as to the transaction between the original owner and the purchaser. Receiving a cold shoulder, the town filed suit in state court.
See id.
at 540.
On those facts, the SJC remanded the case for a determination of the purchaser’s intent at the time he took title.
Id.
at 546. In language celebrated by BTG and its shareholders, the court concluded that “[a]t common law, a right of first refusal ripens into an option to purchase when the condition set forth in the instrument creating the right is met.... The holder is entitled to specific performance of the option as to a subsequent owner who purchased with notice of the holder’s right of first refusal.”
Id.
at 543 (footnote omitted). Although the SJC’s commentary in
Town of Sud-bury
is relatively broad, the case is readily distinguishable from the case at hand on at least two levels: the purpose underlying the right of first refusal and the centrality of the remedy to the litigation.
As to the first, the Massachusetts statute granting the right of first refusal in
Toum of Sudbury
was passed under the aegis of the legislature’s power to regulate “for the purpose of developing and conserving agricultural or horticultural lands.” Mass. Const, amend, art. 99;
see Town of Sudbury,
787 N.E.2d at 544-46. Thus, specific performance embodied the
only
remedy that would comport with the right’s purpose: only by acquiring the land could the town maintain it for agricultural use. In contrast, BTG’s right of first refusal was negotiated by the parties to the Shareholders’ Agreement in order to ensure that Harvard (and not some stranger who did not have BTG’s blessing) would retain control of BTG’s preferred shares. Rescission of the unauthorized transfer returns the shares to a Harvard entity (Holdings) and, thus, adequately serves the original purpose of the right. Specific performance, however, would return the shares to BTG, negating any opportunity for a further relationship between Harvard and the company — a relationship clearly valued at the time of contracting.
The second distinction is equally compelling. In
Toum of Sudbury,
the trial court’s remedial choice was hardly central to the litigation; in fact, the language to which BTG and its shareholders cling so tightly is dictum, pure and simple. That is understandable; the issue actually litigated in
Toum of Sudbury
was whether notice of an intended change of use as opposed to actu
al change of use triggered the town’s statutory first refusal right.
See Town of Sudbury,
787 N.E.2d at 544. The court’s choice of remedy was seemingly uncontroversial. No matter whether the court rescinded the transaction (and, thus, returned the land to the original owner) or specifically enforced the right of first refusal (and, thus, required the land to be offered to the town), the purchaser would lose the opportunity to develop it for a non-agricultural purpose.
Here, however, the consequences of the remedial choice are quite different. Although Holdings and the Fund are not affiliates as that term is defined in the Shareholders’ Agreement, they nonetheless are related entities. As such, their interests are much more closely aligned than those of the original owner and purchaser in
Town of Sudbury.
Holdings and the Fund are not indifferent to the choice of remedy — nor should they be. If rescission stands, the BTG shares will remain in a Harvard pocket, but if specific performance is ordered, the shares (and their apparently enhanced value) will escape completely from the fold.
On the basis of these distinctions, we conclude that the district court committed no error of law when it declined to order specific performance in this instance. Even so, BTG and its shareholders have a fallback position. They asseverate that the district court misapplied Massachusetts law in two other respects. Accordingly, we examine those asseverations.
First, BTG and its shareholders assert that the court’s remedial decision offends basic canons of contract interpretation by transforming a right of first refusal into a mere veto right. This assertion lacks force. Although we agree that courts must enforce contracts as written,
see, e.g., Hakim v. Mass. Insurers’ Insolvency Fund,
424 Mass. 275, 675 N.E.2d 1161, 1164 (1997), the shareholders’ characterization of the district court’s actions is unfair. By ordering rescission, the district court did not alter the underlying contract — if Holdings were to attempt to retransfer the shares to the Fund (or any other non-affiliate, for that matter), it would still need to abide by the right of first refusal.
Second, BTG and its shareholders hypothesize that the district court selected an incorrect remedy. They find fault because the remedy imposed does not place them in the same position that they would have occupied had Holdings performed as required by the right of first refusal. That argument assumes too much.
While it is the goal of specific performance to place parties in the same position as they would occupy had the contract been carried out, that is not the goal of
all
contract remedies.
Cf. VMark Software, Inc. v. EMC Corp.,
37 Mass.App. Ct. 610, 642 N.E.2d 587, 590 n. 2 (1994) (contrasting “[t]he long-established general rule for breach of contract recovery in Massachusetts ... that the wronged party should ... be placed in the same position as if the contract had been performed” with an “alternative” remedy designed to place a plaintiff in as good a position as he would have occupied had no transaction occurred). Rescission is a remedy of a different kind — a remedy that, in this case, aims to place BTG and its shareholders in as good a position as they would have enjoyed had the offending transaction not occurred at all. Given the district court’s superior coign of vantage, we are not persuaded that the court committed legal error in weighing the equities, favoring some remedial interests over others, and eon-
eluding that rescission was the remedy of choice.
An additional circumstance buttresses our view that the district court exercised its judgment reasonably. Even without specific performance, BTG likely will end up with the entirety of its preferred shares through the put, which Holdings has already tried to exercise. Given the incestuous nature of the relationship between Holdings and the Fund — while they are not affiliates within the isthmian confines of the definition embedded in the Shareholders’ Agreement, they are, as we have said, plainly related parties — the put price may well represent a fairer approximation of the worth of the stock at or near the time of the attempted transfer than would the terms agreed upon by Holdings and the Fund.
The BTG shareholders most bruited response to this reasoning is that a weighing of the equities rewrites the district court’s remedial rationale. This position relies heavily on the district judge’s casual statement that he was “not doing equity.” Yet, the judge retreated from that remark almost immediately thereafter, saying “I guess maybe I am specifically enforcing it in the sense that I’m saying that the shares have to be retransferred.” Given that these comments were made from the bench and were not repeated in the Rule 54(b) findings, we deem it wise to embrace substance over form.
See, e.g., United States v. Hilton,
946 F.2d 955, 958 (1st Cir.1991) (“We think it is unrealistic to expect that busy trial judges, ruling from the bench, will be infinitely precise in their choice of language.”). To cinch matters, the judge characterized the transaction as “void
ab inito
” — a characterization more consistent with rescission than with specific performance.
We need go no further. Finding, as wé do, that the district court was not limited to a single remedy under Massachusetts law, that rescission of the transfer was within the armamentarium of permissible choices available to the court, and that rescission was reasonable under the circumstances, we conclude that the court acted within the encincture of its discretion in simply voiding the transfer.
III. CONCLUSION
We summarize succinctly. Holdings and the Fund are not affiliates within the purview of the Shareholders’ Agreement. Thus, Holdings breached that agreement when it did not offer the BTG shareholders an opportunity to exercise their right of first refusal prior to effecting the challenged transfer. Considering all the circumstances of this case, however, the lower court was not obliged to order specific performance of the first refusal right. Rather, the court acted within its equitable discretion when, after mulling both specific performance and rescission, it chose the latter.
The decision of the district court is affirmed in all respects and the case is remanded for further proceedings. All parties shall bear their own costs.