dissenting.

I respectfully dissent from the opinion of my colleagues. I do not agree that Eisenhower and Doubleday intended Crusade in Europe to be a “work-for-hire” under the 1909 Copyright Act. Given the state of copyright law at the time that Eisenhower and Doubleday entered their agreement, I do not believe the parties intended that the work be “for hire” as that term was then understood. Furthermore, the contract between Eisenhower and Doubleday, the tax treatment that Eisenhower received for the manuscript, and the evidence in the record that several publishers were offered the opportunity to compete for the right to publish Eisenhower’s manuscript point to one conclusion only: that Eisen *886 hower sold Doubleday a product, not his services. Thus, I conclude that Crusade in Europe was not written as a work for hire, and accordingly would reverse the district court judgment in favor of Doubleday.

As the majority explains, the 1909 Copyright Act does not contain a definition for the term “work-for-hire.” The majority describes how courts have developed the ’ doctrine establishing the meaning of the term, and how that doctrine has evolved to encompass a broader range of employment relationships. As the majority acknowledges, this court-made doctrine initially held that only works created by traditional “employees” fell under the “work for hire” doctrine. Id.; see also Cmty for Creative Non-Violence v. Reid, 490 U.S. 730, 749 , 109 S.Ct. 2166 , 104 L.Ed.2d 811 (1989) (“CCNV”), Self-Realization Fellowship Church v. Ananda Church of Self-Realization, 206 F.3d 1322, 1327 (9th Cir.2000). It was not until the 1960s that this Circuit for the first time included work created by independent contractors in the scope of the “work for hire” doctrine. CCNV, 490 U.S. at 749 , 109 S.Ct. 2166 ; Lin-Brook Builders Hardware v. Gertler, 352 F.2d 298, 300 (9th Cir.1965). The presumption that a given work is “for hire” is simply that — a presumption, “based on the presumed mutual intent of the parties.” May v. Morganelli-Heumann & Assocs., 618 F.2d 1363, 1368 (9th Cir.1980). I conclude that as the law stood in 1948, it is unlikely that either party to the transaction viewed the sale of Crusade in Europe to Doubleday as the creation of a work for hire. No one has suggested that in 1948, when Eisenhower entered his deal with Doubleday, he was in a traditional employment relationship with Doubleday. Yet the law governing “work for hire” in 1948 extended only to such relationships. Thus, I fail to see how the majority can conclude that the parties intended the work to be “for hire” when that doctrine was understood to encompass only works created by employees.

Furthermore, in making its factual findings, the district court relied heavily on documents and testimony addressing Eisenhower’s and Doubleday’s hindsight reconstruction of the events, to the exclusion of other evidence in the record that cast the events in a different light. For instance, the record contains evidence that Eisenhower had been seeking a publisher prior to his meeting with Doubleday, such as notes by Robinson (of the Tribune), indicating that Eisenhower was looking for a publisher (as opposed to Doubleday looking for an author), as well as evidence that other publishers were offering to buy Eisenhower’s manuscript. The record also contains evidence in the form of correspondence with Dick Simon of Simon & Schus-ter that reveals that Eisenhower had made initial attempts to write his memoirs over a year before his meeting with Robinson. The record shows that in 1947, Eisenhower wrote to his wife in order to convey his papers in the event of his death, revealing that Eisenhower had long contemplated that his writings could be lucrative and take care of his family’s financial needs. It is hardly as clear that “Eisenhower would not have written the work without the urging of Doubleday and Tribune,” as the majority appears to indicate.

The majority simply defers to the district court’s interpretation of the facts. But the district court selectively relied on after-the-fact recollections and the testimony of a witness who was not even employed at Doubleday during the negotiations with Eisenhower. In my opinion, this reliance resulted in a lopsided view of the facts that is overly influenced by the way the parties viewed the events many years down the road, once the significance of the events was clear in the parties’ minds. I cannot arrive at the majority’s *887 conclusion about the parties’ intentions without disregarding significant evidence in the record, and I have no reason to assume that the evidence that has been disregarded is any less reliable than the evidence that the district court relied upon.

The majority concludes that neither the contract for the sale of Crusade in Europe nor the tax treatment of the income that Eisenhower received rebuts the presumption that the parties intended to create a work for hire. Here, again I must part ways with my colleagues. I cannot read the contract for the manuscript’s sale without concluding that it clearly provided for the sale of publication rights, an action that is completely at odds with the idea that the book had been created “for hire.”

The contract states: “I hereby sell, assign, transfer and set over unto Doubleday ... for its own use, absolutely and forever, the manuscript of my War Memoir entitled CRUSADE IN EUROPE, and all rights of every nature pertaining thereto.” The explicit terms of the contract reveal that Eisenhower had the right, at the time of the transaction, to sell his copyright interest in the work to Doubleday. The contract thus reflects a transfer of Eisenhower’s common law copyright interest to Doubleday. See, e.g., Urantia Found’n v. Maaherra, 114 F.3d 955, 960 (9th Cir.1997) (“Under the 1909 Copyright Act, an unpublished work was protected by common law copyright from the moment it was created ... ”) (citation omitted). However, if Crusade in Europe were a work for hire, the copyright would have vested in Doubleday from the moment it was created. See, e.g., Dolman, 157 F.3d at 713. In Dolman, we considered the copyright to songs created by Roy Shield, who was an employee of Victor Talking Machine. 157 F.3d at 710. The composer assigned his copyright in the songs to Southern Music Publishing Company, the publishing arm of Victor. Id. We concluded that the assignment of the copyright rebutted the presumption that the employee’s work was for hire. “Had the works been intended to be works for hire for Victor, there would have been no reason for Southern to accept an invalid assignment of rights from Shield, knowing that its parent company already owned those rights.” Id. at 713. This sentiment is equally relevant here— why would Eisenhower transfer his copyright to Doubleday if it was already in Doubleday’s name by virtue of its status as being created at Doubleday’s “instance and expense”? 1 If the oral agreement of the parties was intended to transfer ownership of the manuscript to the Doubleday prior to its completion, the signed agreement would have no purpose. Eisenhower would have had nothing to convey to Doubleday: he would have conveyed the rights to the manuscript by virtue of its creation as a “work for hire.” I conclude that the contract on its face rebuts the presumption that the parties intended the work to be for hire.

Similarly, I am not persuaded by the majority’s analysis of the tax treatment of the proceeds of the sale of Crusade in Europe. In my opinion, the tax treatment that Eisenhower obtained rebuts the presumption that he was “hired” to write a book. Although tax law does not dictate the copyright status of a work, it reveals how the parties themselves viewed their employment arrangement. In this case, *888 the income was treated as if it were derived from the sale of an asset, not from personal income. Clearly, Eisenhower was not paid by Doubleday for his writing services; he was paid by Doubleday for the end product.

I cannot look at the record in this case without concluding that these facts are far from “the prototypical situation where the motivating factor for producing the work was the employer’s inducement.” In my mind, it is clear that Eisenhower wanted to write Crusade in Europe, and would have done so under the auspices of the publisher who arranged things most favorably to him. By calling Crusade in Europe a “work for hire,” the majority establishes a standard by which nearly any work produced with the financial and logistical support of a publisher could be considered to have been produced “for hire.” .1 do not think that such an expansive conception of the term was intended by the drafters of the 1909 Copyright Act. Therefore, I must dissent.

. The majority attempts to distinguish Dol-man by pointing out that in that case, the assignee of the copyrights named the composer as the author in copyright registrations. I fail to see how that fact is relevant, and note that in this case, Doubleday specified that Eisenhower was the author of the work on its initial copyright registration. Doubleday did not indicate that the work was “for hire” until it filed a copyright renewal in 1975.