Milla Karkkainen filed a Petition for Return of Child under the Hague Convention on the Civil Aspects of International Child Abduction, Oct. 25, 1980, T.I.A.S. No. 11670, 19 I.L.M. 1501 ("Hague Convention"). Karkkainen alleged that her ex-husband, Vladimir Kovalchuk, and his current wife, Julie d'Itri (collectively, the "Respondents"), wrongfully retained her daughter, Maria Kovalchuk, when she was eleven years old. The District Court denied the petition, holding that there was no wrongful retention because Maria's habitual residence was the United States. Karkkainen appeals this decision, arguing that Maria is habitually resident in Finland. Although this is a close case, we believe that, prior to her retention, Maria acclimatized to the United States and that there was a degree of settled purpose from her perspective to remain in this country. The existence of shared parental intent to permit Maria to choose her country of residence bolsters this conclusion. Thus, we agree with the District Court's finding that Maria is a habitual resident of the United States and will affirm.

Maria was born on April 25, 1992 in Russia. Her parents, Milla Karkkainen and Vladimir Kovalchuk, were married at the time and remained so until 1997. After their divorce, Karkkainen and Kovalchuk agreed that Maria would live with her mother in Finland.

Both Karkkainen and Kovalchuk remarried after their split. Karkkainen married Kimmo Karkkainen in January 1998, and Kovalchuk married Julie d'Itri in September 2000. Kovalchuk and d'Itri saw Maria periodically, either by traveling to Europe or by bringing her for visits to the United States. In 2000, however, Maria was unable to obtain a tourist visa for a visit to the United States due to the concerns of the United States Consulate in Finland about Maria's custody status.

As part of their effort to obtain a visa for Maria to visit the United States, Kovalchuk and Karkkainen signed a Stipulation in Custody in December 2000 that clarified their custody arrangement. The Stipulation provided, inter alia, that Karkkainen "shall have primary physical custody of [Maria], including the right of the child's residence in Finland, which for purposes of The Hague Convention on the Civil Aspects of Child Abduction, shall be considered the child's `habitual residence.'" (Stipulation in Custody ¶ 5 at App. 503.) The parents continued to share legal custody of Maria such that both had "the right and responsibility to make major decisions affecting . . . [her] best interest." (Stipulation in Custody ¶ 3 at App. 503.) The terms of the Stipulation required that it be filed with the Court of Common Pleas of Allegheny County, Pennsylvania. Though the Stipulation was signed by a common pleas judge, the parties never actually filed it with the court. The record reflects that the primary purpose of the Stipulation in Custody was to secure Maria's tourist visa.

Despite these efforts to clarify Maria's custody status, Maria was unable to obtain a visa to visit the United States. In February 2002, the parties began discussions about making Maria a permanent resident of the United States for immigration purposes, which would dispense with the need for a tourist visa. In March 2002, Karkkainen granted permission for Maria to become a permanent American resident in a signed and notarized document. It was Karkkainen's understanding at that time that she would not lose any custody rights over her daughter if she were made a permanent resident of the United States, but that the change in immigration status would give Maria the right to remain in the United States indefinitely. Maria was granted an immigrant visa in September 2002 and she officially became a permanent resident of the United States when she visited the country in October 2002. She also visited the United States in December 2002 during the Christmas holiday and over her Easter break in April 2003.

During the winter of 2002 and spring of 2003, Maria, her parents, and her stepparents began to plan for Maria to make a longer visit to the United States. The parties agreed that she would spend the entire summer here with her father and d'Itri. In addition, Maria began increasingly to express her preference to move permanently to the United States. Maria had a conversation in May 2003 with her mother and stepfather in which she stated that she wanted to live with her father. Maria's stepfather told her that she was free to make that decision. When Karkkainen did not disagree with this statement, Maria was left with the impression that she had been given permission to move permanently to the United States if she wished. After this conversation, Maria said goodbye to her teacher, Tuula Merenheimo, and to several friends, telling them that she was moving to the United States. As a parting gift, Merenheimo gave Maria the books that she would have used during the next school year in Finland. These books were usually kept by the teacher during the summer and handed out at the beginning of the academic year.

Several events reinforced Maria's belief that she would be permitted to move to the United States permanently. Milla and Kimmo Karkkainen helped Maria apply to a private American school for the fall semester of 2003 by faxing her academic transcripts to the school. Maria heard her mother tell her grandmother on the telephone that Maria was moving to the United States. And Karkkainen let Maria travel to the United States on June 6, 2003, moments after Maria told Karkkainen that she was unsure she would return to Finland at the end of the summer.

The central factual dispute of this case is what the understanding of the parties was at the time Maria came to the United States. The Respondents claim that the parties agreed that the summer would be a trial period during which Maria would decide whether she wanted to move to the United States permanently. Karkkainen argues that she never granted permission for Maria to live in the United States indefinitely and that she expected Maria to return to Finland on August 10, 2003.

The record reflects that Maria is both mature and intelligent for her age. An expert in teaching and training children in the performing arts testified that Maria is "a very focused, gifted, talented and . . . creative child" with particularly strong skills in photography and drawing. An independent child psychologist found that Maria was "uniquely talented and highly intelligent," an impression the District Court echoed after hearing Maria's testimony. Maria could communicate well in Finnish, English, and Russian, and had extensive experience traveling in Europe and the United States for visits with her father. She was, in short, much more experienced and mature than the average eleven year old when she came to the United States on June 6, 2003.

During the summer of 2003, Maria took academic classes, studied photography, traveled in the United States, and cultivated her relationships with d'Itri and d'Itri's family. In addition, Maria was admitted to a private American school named The Ellis School, where she enrolled to attend in the fall. When Maria did not return to Finland in August 2003, Karkkainen filed a Petition for Return under the Hague Convention. The District Court determined that Maria had become acclimatized during her stay in the United States prior to the date of her retention. Accordingly, the Court found that Maria was a habitual resident of the United States and refused to return her to Finland.

The District Court had subject matter jurisdiction under 28 U.S.C. § 1331 and 42 U.S.C. § 11603(a). We have appellate jurisdiction under 28 U.S.C. § 1291.

II. The Legal Framework of the Hague Convention

The two main purposes of the Hague Convention are "to ensure the prompt return of children to the state of their habitual residence when they have been wrongfully removed," Hague Convention, pmbl., and "to ensure that rights of custody and of access under the law of one Contracting State are effectively respected in the other Contracting States," id., art. 1. The Convention's procedures are not designed to settle international custody disputes, but rather to restore the status quo prior to any wrongful removal or retention, and to deter parents from engaging in international forum shopping in custody cases. Baxter v. Baxter, 423 F.3d 363, 367 (3d Cir.2005).

Any person seeking the return of a child in the United States may commence a civil action under the Convention by filing a petition in a court of the jurisdiction in which the child is located. 42 U.S.C. § 11603(b). To obtain an order for the child's return under the Hague Convention, the petitioner bears the burden of proving by a preponderance of the evidence that the removal or retention was wrongful under Article 3. 42 U.S.C. § 11603(e)(1)(A). Under Article 3 of the Hague Convention, the removal or retention of a child is "wrongful" where:

a. it is in breach of rights of custody attributed to a person, an institution or any other body, either jointly or alone, under the law of the State in which the child was habitually resident immediately before the removal or retention; and

b. at the time of removal or retention those rights were actually exercised, either jointly or alone, or would have been so exercised but for the removal or retention.

Hague Convention, art. 3.

A petitioner cannot claim that the removal or retention of a child is "wrongful" under the Hague Convention unless "the child to whom the petition relates is `habitually resident' in a State signatory to the Convention and has been removed to or retained in a different State." Gitter v. Gitter, 396 F.3d 124, 130 (2d Cir.2005) (emphasis added); see also Miller v. Miller, 240 F.3d 392, 398 (4th Cir.2001) (requiring petitioner to prove that children were habitually resident in a country other than the one to which they were removed). Determination of a child's habitual residence immediately before the alleged wrongful removal or retention is therefore a threshold question in deciding a case under the Hague Convention. Feder v. Evans-Feder, 63 F.3d 217, 222 (3d Cir. 1995).

Thus, we have noted that wrongful removal or retention claims under Article 3 of the Convention typically raise four questions: (1) When did the removal or retention at issue take place? (2) Immediately prior to the removal or retention, in which state was the child habitually resident? (3) Did the removal or retention breach the rights of custody attributed to the petitioner under the law of the habitual residence? (4) Was the petitioner exercising those rights at the time of the removal or retention? See Baxter, 423 F.3d at 368 (citing Mozes v. Mozes, 239 F.3d 1067, 1070 (9th Cir.2001)). Because the parties stipulated that Karkkainen had rights of custody under Finnish law that she was exercising prior to Maria's retention, only the first two issues are in contention on this appeal.

Even when a court finds wrongful removal or retention, it is not necessarily required to return a child to its habitual residence. After a petitioner demonstrates wrongful removal or retention, the burden shifts to the respondent to prove an affirmative defense against the return of the child to the country of habitual residence. Baxter, 423 F.3d at 368; see also Hon. James D. Garbolino, International Child Custody Cases: Handling Hague Convention Cases in U.S. Courts ch. 5 (3d ed.2000) (discussing the affirmative defenses under the Convention). These affirmative defenses are narrowly construed to effectuate the purposes of the Convention and, even where a defense applies, the court has the discretion to order the child's return. See Feder, 63 F.3d at 226; Hague International Child Abduction Convention, Text and Legal Analysis, 51 Fed.Reg. 10,494, 10,509 (Mar. 26, 1986).

III. The District Court's Evidentiary and Factual Rulings

Though the fundamental issue in this appeal is whether the District Court correctly determined Maria's habitual residence, Karkkainen also contends that the District Court erred in several evidentiary rulings and in its fact-finding. Because these rulings formed the basis for the District Court's holding that Maria was not wrongfully retained, we take them up before turning to the merits of Karkkainen's claim under Article 3 of the Convention.1

We review the District Court's determinations concerning the admissibility of evidence for abuse of discretion. Forrest v. Beloit Corp., 424 F.3d 344, 349 (3d Cir.2005). To the extent an evidentiary issue turns on the interpretation of a Federal Rule of Evidence, rather than the mere application of the rule, our review is plenary. Id.

Karkkainen claims that the District Court abused its discretion by appointing an expert to evaluate Maria's level of maturity that lacked sufficient experience in "parental alienation syndrome." Karkkainen argues that one of the central themes of her case was that Maria had been alienated from her mother by the Respondents and that the District Court therefore should not have considered Maria's desire to move to the United States permanently. As evidence of Maria's alienation, Karkkainen cites an instance in which Maria called her mother "aunt." However, as the District Court noted, Karkkainen raised the claim of Maria's alienation only as a way to rebut the Respondents' affirmative defenses. Because the District Court ruled that the United States was Maria's habitual residence, Respondents' affirmative defenses were not at issue and any lack of knowledge on the part of the expert about "parental alienation syndrome" was irrelevant.

The District Court heard testimony from two witnesses that Karkkainen argues should not have testified. The custodian of records from The Ellis School testified as to whether the school relied on documents sent by Maria's Finnish teacher for her admission. Though Karkkainen contends that this testimony constituted unfair surprise, the records custodian was listed as a witness on the witness list. The District Court did not abuse its discretion in permitting her to testify as to factual matters about which she had personal knowledge. Karkkainen also argues that Maria should not have been permitted to testify. However, the District Court held, based on expert evidence, that Maria was mature enough to testify. Because Maria's perspective was central to the question of her habitual residence under the Hague Convention, see Feder, 63 F.3d at 224 (defining a child's habitual residence in terms of the child's perspective), it was clearly appropriate for the District Court to admit her testimony.

The District Court admitted testimony about whether Maria was "well-settled." Karkkainen argues that the "well-settled" defense of Article 12 of the Hague Convention is inapplicable in this case and that the District Court should have rejected testimony as to this issue. The District Court reserved judgment on whether the well-settled defense applied and stated that it would consider the testimony only if the defense later became relevant. There is no evidence that the District Court used such testimony to reach its decision on habitual residence, the only subject of its holding. Likewise, we reject Karkkainen's claim that the District Court improperly permitted testimony regarding Maria's best interests. Karkkainen points to no specific instances in which the District Court permitted such testimony, and we have found none within the record. We also conclude that the District Court admitted hearsay testimony only under the exceptions of the Federal Rules and properly limited its use. Thus, we find no abuse of discretion on these points.

Finally, Karkkainen argues that the District Court erred in finding that the parties agreed to allow Maria to choose whether she would live in the United States indefinitely or return to Finland at the end of the summer of 2003. We must review this factual finding for clear error. Baxter, 423 F.3d at 367; Delvoye v. Lee, 329 F.3d 330, 332 (3d Cir.2003). We will not reverse as long as the District Court's account of the evidence is "plausible in light of the record," even if convinced that we "would have weighed the evidence differently." Anderson v. City of Bessemer, 470 U.S. 564, 574, 105 S. Ct. 1504, 84 L. Ed. 2d 518 (1985).

The record supports the District Court's conclusion that the parties agreed when Maria left for the United States in June 2003 that she would choose where she would reside after the summer. At an interview at the American Embassy in Helsinki in September 2002, d'Itri explained to a consular officer that Maria would have the option of remaining in the United States after her visit in the summer of 2003. Karkkainen was present during the conversation and did not object to the proposed plan. Email exchanges and multiple discussions between the parties leading up to Maria's 2003 trip to the United States also suggested that Karkkainen agreed Maria was "free to go" to the United States indefinitely. In addition, Karkkainen aided the Respondents in their efforts to place Maria in The Ellis School for the Fall 2003 school semester and never contacted the school to indicate that she did not wish Maria to attend. Indeed, on the very day Maria left for the United States, she and her mother discussed the possibility that Maria would not return to Finland at the end of the summer. Taken together, this and other evidence in the record supports the District Court's finding that Karkkainen, the Respondents, and Maria all understood that Maria would have the choice of remaining in the United States and that she would not necessarily return to Finland in August 2003.

At the same time, there is evidence in the record that could support a different conclusion. For example, Karkkainen never expressly gave permission for Maria to attend school in the United States. Moreover, the Respondents stated in a January 2003 email that it was their understanding that Maria would return to Finland in August 2003. The fact that Maria had a round-trip ticket with which she could return to Finland on August 10, 2003 may also suggest that the parties did not intend Maria to remain in the United States indefinitely. Despite such evidence, we will not set aside the District Court's finding given our deferential clear error standard.

Having resolved Karkkainen's evidentiary and factual objections, we must determine when the alleged wrongful retention occurred so as to establish the relevant date of Maria's habitual residence for purposes of the Hague Convention. The District Court held that the date of retention was August 28, 2003, the date on which Karkkainen filed her petition for Maria's return. Prior to that time, the Court found that there was an ongoing dispute between the parties about whether the agreement to allow Maria to stay permanently in the United States remained in effect. Once Karkkainen filed the petition for Maria's return, she unequivocally signaled her opposition to Maria's presence in the United States. After that date, there was no doubt that Maria remained with her father against her mother's wishes and was therefore retained. See Slagenweit v. Slagenweit, 841 F. Supp. 264, 270 (N.D.Iowa 1993) ("The wrongful retention does not begin until the noncustodial parent. . . clearly communicates her desire to regain custody and asserts her parental right to have [her child] live with her." (emphasis added)).

Karkkainen argues that August 10, 2003, not August 28, 2003, is the proper date of retention. This was the date of Maria's return airline ticket to Finland and the date after which Kimmo Karkkainen indicated in a July 2003 email to the Respondents that Maria's presence in the United States would constitute kidnapping. Karkkainen argues that there is no legal basis for setting the date of retention as the day on which she unequivocally communicated her opposition to Maria's presence in the United States and that, in any event, she had clearly communicated her opposition prior to August 28, 2003.

This case does not require us to decide whether a child is not retained under the Convention until a parent unequivocally communicates his or her desire to regain custody. We assume that this standard applies, but hold that it was clearly erroneous for the District Court to find that Karkkainen had not clearly communicated her opposition to Maria's presence in the United States until she filed the petition for return. There is unrebutted evidence in the record showing that, by mid-July 2003, Karkkainen had withdrawn her consent to have Maria remain in the United States beyond August 10, 2003 and that the Respondents were fully aware of this. Neither the District Court nor the Respondents pointed to anything in the record that suggests there was confusion about Karkkainen's opposition after mid-July 2003, and we have found no such evidence ourselves. Under these circumstances, we must set aside the District Court's factual finding and accept as the date of retention August 10, 2003, prior to which it is undisputed that Maria was present in the United States with Karkkainen's permission. See Toren v. Toren, 191 F.3d 23, 27-28 (1st Cir.1999) (finding no retention where child's presence in United States was consistent with agreement of parents).

Whether the Respondents wrongfully retained Maria under the Hague Convention will be determined by where Maria habitually resided immediately prior to her alleged wrongful retention on August 10, 2003. If we find Finland was Maria's habitual residence on that date, we must also find that her retention in the United States was wrongful; if we conclude that Maria was a habitual resident of the United States, her retention here would not be wrongful under Article 3 of the Convention. See Gitter, 396 F.3d at 130; Miller, 240 F.3d at 398; Feder, 63 F.3d at 222. The determination of a child's habitual residence presents a mixed question of fact and law. Whiting v. Krassner, 391 F.3d 540, 546 (3d Cir.2004). We therefore "review the [D]istrict [C]ourt's underlying findings of historical and narrative facts for clear error, but exercise plenary review over the court's application of legal precepts to the facts." Delvoye, 329 F.3d at 332.

The Convention does not specifically define the term "habitual residence." Though the Courts of Appeals have employed slightly different tests for habitual residence, each test has in common the goal of determining where a child's home is at the time of removal or retention. These tests facilitate the primary objective of the Hague Convention: to ensure stability in a child's family and social environment. See Elisa Perez-Vera, Explanatory Report ¶¶ 11, 24, 72, in 3 Hague Conference on Private International Law, Acts and Documents of the Fourteenth Session, Child Abduction 426 (1982) ("Perez-Vera Report").2

The inquiry into a child's habitual residence is a fact-intensive determination that cannot be reduced to a predetermined formula and necessarily varies with the circumstances of each case. Whiting, at 546. This is especially true in cases such as this one, where the petitioning parent initially agreed to allow the child to stay abroad for an indefinite duration, but subsequently had second thoughts about that decision. Mozes, 239 F.3d at 1077. "These cases . . . generally have no clear answer and are very fact-dependent." Whiting, 391 F.3d at 549.3

We have stated that a child's habitual residence is "the place where he or she has been physically present for an amount of time sufficient for acclimatization and which has a `degree of settled purpose' from the child's perspective." Feder, 63 F.3d at 224. This approach considers a child's experience in and contacts with her surroundings, focusing on whether she "develop[ed] a certain routine and acquire[d] a sense of environmental normalcy" by "form[ing] meaningful connections with the people and places [she] encountered" in a country prior to the retention date. Id. at 550-51. We examine a child's conduct and experiences to determine whether she became "firmly rooted" in her new surroundings, not merely whether she acculturated to a country's language or customs. Holder v. Holder, 392 F.3d 1009, 1019 (9th Cir.2004); see also Mozes, 239 F.3d at 1078-79 (describing acclimatization as being "firmly embedded in the new country" or "being well-adjusted in one's present environment"). Thus, if a child becomes rooted in one country, we will not return her to another one where doing so would take her "out of the family and social environment in which [her] life has developed." Perez-Vera Report ¶ 11. Simply put, this inquiry considers whether a child has made a country her home before the date of her removal or retention. See Holder, 392 F.3d at 1019.

Though we examine acclimatization and settled purpose "from the child's perspective," Feder, 63 F.3d at 224, we consider parental intent as part of this inquiry "because the child's knowledge of these intentions is likely to color its attitude to the contacts it is making," Mozes, 239 F.3d at 1079-80. See also Silverman v. Silverman, 338 F.3d 886, 898 (8th Cir.2003) (noting that, although courts must focus on the child, "parental intent is also taken into account"). As the Court of Appeals for the Ninth Circuit noted, the intentions of a child's parents may affect the length of time necessary for a child to become habitually resident or otherwise influence a child's ability to acclimatize. Mozes, 239 F.3d at 1079-80.

In addition to considering how parental intent affected a child's perspective, we must also give some independent weight to "the parents' present, shared intentions regarding their child's presence" in a particular place. Feder, 63 F.3d at 224 (emphasis added). This approach helps courts ensure that neither parent is acting unilaterally to alter a joint understanding reached by the parents. Thus, given the facts of this case, the agreement between Maria's parents is relevant in two ways. First, it affects the amount of time required for Maria to become acclimatized and to demonstrate a degree of settled purpose to live in the United States. Second, the agreement is relevant on its own under the shared intent inquiry.

It is obvious based on these standards that Maria was a habitual resident of Finland prior to her arrival in the United States on June 6, 2003. She lived in that country for years and was acclimatized to her environment there in every way. In addition, Finland was the country that Maria's parents had set as her habitual residence in their Stipulation in Custody. The question before us, then, is whether Maria's habitual residence changed from Finland to the United States prior to the retention date.

Before applying these standards to the instant case, we address Karkkainen's claim that the Stipulation in Custody prevented a change in Maria's habitual residence, as it remained binding and rendered Maria habitually resident in Finland on the retention date. The District Court disagreed with this argument and found that the parties had modified the Stipulation by developing a shared intention that Maria would be permitted to decide during the summer of 2003 whether she wished to move to the United States or return to Finland. Because we will not set aside the District Court's finding that the parties agreed Maria would decide for herself whether to return to Finland, we must determine whether this shared intention effectively modified the Stipulation in Custody.

Though it was signed by a common pleas judge, the Stipulation in Custody was never filed with the Court of Common Pleas of Allegheny County, as its terms required. Consequently, it is an informal custody agreement under Pennsylvania law, not a binding court order. Witmayer v. Witmayer, 320 Pa.Super. 372, 467 A.2d 371, 374 (1983). While a court will generally look favorably upon an informal custody agreement, "if circumstances change after the making of the agreement. . . , then lesser or no weight should be afforded" to it. Id. at 375. The District Court believed that the shared intention of the parties that Maria would choose where to live constituted a changed circumstance that modified the Stipulation in Custody. We agree. The shared intention of Maria's parents changed the underlying assumption of the Stipulation when it was prepared, namely, that Maria would necessarily live with her mother. Once the parties reached a new agreement about how Maria's residence would be determined and gave her the option to live in the United States with her father, circumstances had changed and the habitual residence term of the Stipulation in Custody was no longer binding.

Furthermore, the Stipulation was prepared for the express purpose of obtaining a tourist visa for Maria. Maria has since become a permanent legal resident of the United States, making a tourist visa unnecessary for her to visit or remain in this country. Because Maria's immigration status was critical to the factual context in which the parties drafted the Stipulation, this change buttresses our conclusion that the instant case involves different circumstances from those present when the parties entered the Stipulation and that the Stipulation's lasting effectiveness is suspect. Id. We therefore afford no weight to the provision of the Stipulation in Custody that sets Finland as Maria's habitual residence.

C. Acclimatization and Degree of Settled Purpose

To determine whether Maria's habitual residence changed from Finland to the United States, we first consider whether she acclimatized to the United States prior to the date of retention and whether there was a degree of settled purpose from her perspective to remain in this country beyond August 10, 2003. Whiting, 391 F.3d at 550. Courts have identified a number of specific factors that are indicative of acclimatization and a degree of settled purpose from the child's perspective. In Feder, we noted that academic activities are among "the most central . . . in a child's life" and therefore highly suggestive of acclimatization. 63 F.3d at 224. The Court of Appeals for the Eleventh Circuit has taken school attendance, social engagements, and lessons to be evidence of acclimatization. Ruiz v. Tenorio, 392 F.3d 1247, 1255 (11th Cir.2004). The Ninth Circuit Court of Appeals weighed a child's participation in sports programs and excursions in his new country in favor of acclimatization. Holder, 392 F.3d at 1020.

Applying the lessons of such cases to the circumstances before us, there is evidence in the record that Maria acclimatized herself to the United States during the summer of 2003. She enrolled in The Ellis School and took summer classes to prepare for her attendance there in the fall. She also took photography classes that summer, traveled in the country, and developed relationships with d'Itri and her family that she had established during previous visits to the United States in October 2002, December 2002, and April 2003. We view these events in the context of record evidence that Maria is "uniquely talented and highly intelligent," an experienced traveler with strong English skills, and mature for her age. Taken together, these factors suggest that Maria "form[ed] meaningful connections with the people and places she encounter[ed]" in the United States and was therefore acclimatized prior to the date of her retention. Whiting, 391 F.3d at 551.

Furthermore, there is evidence in the record that Maria abandoned Finland as her habitual residence. When Maria came to the United States in June 2003, she brought more personal belongings with her than usual, in anticipation that she would remain here after the summer. See Silverman, 338 F.3d at 898 (noting that transfer of personal possessions indicates settled purpose to remain in new country). Maria's decision in July 2003 to remain in the United States, which she communicated to her parents and stepparents, was essentially a choice to abandon Finland as her habitual residence. "[W]hile our jurisprudence on habitual residency. . . has not heretofore enunciated a need for an intent to abandon a former habitual residency in order to establish a new one, it does seem implicit in the concept of acquiring a new `habitual' residence that the previous `habitual' residence has been left behind or discarded." Whiting, 391 F.3d at 550.

There are also factors that weigh against a finding of acclimatization. "Habitual residence may only be altered by a change in geography and passage of time," Silverman, 338 F.3d at 898, and is a concept that focuses on past experience, not future intentions, Friedrich v. Friedrich, 983 F.2d 1396, 1401 (6th Cir.1993). It is fair to ask whether Maria was physically present in the United States for an amount of time sufficient to provide the experiences required to acclimatize to a new country. Maria visited the United States for less than three months in the summer of 2003. Even a mature eleven-year old may not be able to acclimatize to a new country in such a short period of time, especially since Maria did not decide until mid-July 2003 that she wanted to stay permanently with her father in the United States.

As stated above, the intentions of a child's parents "affect[] the length of time necessary for a child to become habitually resident, because the child's knowledge of these intentions is likely to color its attitude to the contacts it is making." Mozes, 239 F.3d at 1079-80. In this case, Maria's parents intended when Maria left for the United States in June 2003 that Maria would be able to choose where she would live at the end of the summer and that her choice would be respected. This agreement, unusual in cases under the Hague Convention, provided the defining context for Maria's visit in the summer of 2003.

There is considerable evidence that the agreement between Maria's parents "colored her attitude" towards her visit to the United States during the summer of 2003. Maria thought when she arrived that she would be permitted to choose, before the end of the summer, the country in which she would reside permanently. In multiple conversations with parents and stepparents, Maria expressed her desire to pick her residence. She indicated to her friends and teacher prior to leaving Finland that her parents would allow her to choose where she would live after the summer of 2003. Furthermore, Maria told her mother on the day that she was leaving for the United States that she was not sure she would return. The fact that Maria believed that she controlled her own destiny influenced her entire experience in the United States prior to her retention here. Indeed, Maria strongly suspected even before she arrived that she would choose to remain here. Under these circumstances, less time was required for Maria to acclimatize and demonstrate a degree of settled purpose to stay in the United States than would normally be the case.

We are presented with a unique fact pattern, in that Maria's parents agreed in June 2003 that she possessed "the material and psychological wherewithal" to decide where she would reside. Mozes, 239 F.3d at 1076. Any eleven year-old girl could conceivably tell her mother that she prefers to remain in a country where she is visiting, yet this would not necessarily alter her habitual residence. Here, Maria's actions during the summer of 2003, and her declaration that she would remain with her father thereafter, are especially indicative of her settled purpose in light of her parents' agreement to permit her to decide to remain in the United States permanently and to respect whatever choice she made. Viewed in the context of this understanding, Maria's conduct, including her preparatory academic work, enrollment in an American school, and her efforts to connect with d'Itri's family, reflects her acclimatization and a degree of settled purpose to remain more clearly than it would if considered in isolation.

Karkkainen argues that the agreement among the parties was nullified once she withdrew her consent that Maria could remain with her father beyond August 10, 2003, which the record shows occurred in mid-July. One could hardly question the proposition that Karkkainen's protests changed the intent of the parties as it existed in June 2003, but this misses the point. "Habitual residence is intended to be a description of a factual state of affairs, and a child can lose its habitual attachment to a place even without a parent's consent." Mozes, 239 F.3d at 1081. Karkkainen would have us ignore the fact that an understanding existed between the parents at the outset in evaluating Maria's perspective. We cannot do so because this understanding was at the heart of Maria's experiences in the United States. We do not view the agreement that existed in June 2003 as itself determinative of Maria's habitual residence, but rather as important in our consideration as to how Maria experienced her contacts in the United States and, consequently, as a factor that influenced the amount of time required for Maria to acclimatize and demonstrate a degree of settled purpose to remain with her father.

Though the relatively short period of time Maria was present in the United States makes it a close question, we hold that Maria was acclimatized to the United States on the date of her retention and that her conduct demonstrated a degree of settled purpose to remain here. We base this conclusion on the specific facts and circumstances before us, Whiting, 391 F.3d at 546, and rely heavily on Maria's maturity and intelligence, her development of relationships with family and friends in the United States prior to her retention, and her academic work during the summer of 2003 to prepare for attendance at The Ellis School. Importantly, we view these factors against the backdrop of the agreement between Maria's parents to permit Maria choose her own residence. We are mindful that we should avoid setting the bar for acclimatization too low, lest we create an incentive for a parent to remove or retain a child in the hope that the child will quickly acclimatize and not be returned. See id. at 551 (recognizing the problematic incentives in Hague Convention cases); Mozes, 239 F.3d at 1079 ("The greater the case with which habitual residence may be shifted without the consent of both parents, the greater the incentive to try."). However, we are satisfied that in the unique circumstances of this case, Maria's experiences in the United States prior to her retention crossed the line that demarcates acclimatization and indicate a degree of settled purpose from her perspective.

In addition to considering habitual residence from Maria's perspective, Feder requires that we give independent weight to her parents' shared intent. 64 F.3d at 224. We have held that, in cases involving very young children, "the shared intent of the parents in determining the residence of their children [is] of paramount importance" and acclimatization is secondary. Whiting, 391 F.3d at 550. Acclimatization is an ineffectual standard by which to judge habitual residence in such circumstances because the child lacks the ability to truly acclimatize to a new environment. See id. at 550-51 (considering the shared intentions of parents where child is too young to "form meaningful connections with the people and places he encounters each day"); Delvoye, 329 F.3d at 334 (determining habitual residence of a newborn baby through analysis of parents' shared intent). Thus, shared parental intent that a very young child will reside in a new country, even for a limited period of time, is sufficient to establish the child's habitual residence in that country. Whiting, 391 F.3d at 549; Feder, 63 F.3d at 223.

We give somewhat less weight to shared parental intent in cases involving older children, like Maria, who have reached an age where they are capable of becoming "firmly rooted" in a new country. Holder, 392 F.3d at 1019. In such cases, our attention generally turns first to the child's perspective, not the parents' intent. See Feder, 63 F.3d at 224 (noting that the test for acclimatization and degree of settled purpose "must focus on the child"); see also Silverman, 338 F.3d at 898 ("The court should have determined the degree of settled purpose from the children's perspective."); Friedrich, 983 F.2d at 1401 ("To determine the habitual residence, the court must focus on the child, not the parents . . . ."). But shared parental intent remains relevant to habitual residence in all cases under the Hague Convention. The Convention "is designed to restore the `factual' status quo which is unilaterally altered when a parent abducts a child and aims to protect the legal custody rights of the non-abducting parent." Feder, 63 F.3d at 221 (emphasis added); see also Mozes, 239 F.3d at 1079 ("The function of a court applying the Convention is . . . to determine. . . whether one parent is seeking unilaterally to alter the status quo with regard to the primary locus of the child's life."). When the parents share an intent as to the child's habitual residence, it must be given some weight. Were a court to exclude shared parental intent entirely from the habitual residence inquiry, and instead focus solely on a child's contacts and experiences, it would fail to consider whether a parent is acting unilaterally to alter what was jointly intended or agreed upon. Factoring shared parental intent into habitual residence therefore serves one of the primary goals of the Hague Convention.

Our cases have not established with any precision how a court should balance shared parental intent against evidence of acclimatization and settled purpose from the child's perspective where the child is not very young. As the Court of Appeals for the Ninth Circuit noted in Mozes, perhaps the most difficult question in this regard is "when evidence of acclimatization should suffice to establish a child's habitual residence" in the face of "uncertain or contrary parental intent." 239 F.3d at 1078. Proceeding from the premise that it is the parents, not the child, who decide where a child lives, the Mozes Court concluded that, in the absence of settled parental intent, courts should be "slow to infer" merely from a child's contacts with a new country that he or she has abandoned an earlier habitual residence. Id. at 1079. "[A] child can lose its habitual attachment to a place even without a parent's consent," but only when "the objective facts point unequivocally" to the conclusion that a child's relative attachments to two countries have changed. Id. at 1081 (internal quotations omitted). Thus, Mozes established a presumption that shared parental intent (or lack thereof) regarding a change in habitual residence generally trumps evidence of acclimatization. See Gitter, 396 F.3d at 134 (adopting the analysis of Mozes and noting that only in "relatively rare circumstances" would evidence of acclimatization outweigh parental intent); Holder, 392 F.3d at 1020 (requiring showing of acclimatization that "overcome[s] the lack of shared parental intent to abandon. . . the children's habitual residence"). The Mozes Court adopted this approach in order to minimize the incentives for unilateral action on the part of either parent, Mozes, 239 F.3d at 1079, and because it believed that "[c]hildren . . . normally lack the material and psychological wherewithal to decide where they will reside," id. at 1076.

Here, because both shared parental intent and acclimatization support a finding that Maria was habitually resident in the United States, we need not decide how we would weigh these factors against each other if they conflicted, which was the focus of Mozes. Accordingly, we note only that, consistent with our instruction in Feder, courts must consider "the parents' present, shared intentions" as part of the habitual residence analysis. 63 F.3d at 224.

When a removal or retention is in accordance with the shared intent of both parents, there is no unilateral action, and therefore no harm for a court to remedy under the Hague Convention. See Toren, 191 F.3d at 28 ("[T]he children's mere presence in the United States cannot constitute a retention because it is entirely consistent with the parties'. . . agreement."). There is record evidence from as early as September 2002 of shared parental intent to permit Maria to move permanently to the United States. Only as it became clear that Maria would choose to remain with her father did Karkkainen retract her consent to allow Maria to choose her residence. Yet, by that time, Maria had already begun to settle in the United States in reliance on the agreement that she would be permitted to remain here permanently. Thus, the mutual understanding during the summer of 2003 was that Maria would have the opportunity to remain in the United States permanently (that is, to live here with a degree of settled purpose). Karkkainen's change of heart in July 2003 is the type of unilateral act that the Hague Convention was designed to prevent, namely, one that disrupts a child's residential environment and thwarts shared parental intent. We therefore give weight to the agreement that existed when Maria left Finland and not to Karkkainen's subsequent retraction of consent.

In summary, though we find it to be a close question, we agree with the District Court's holding that Maria acclimatized to the United States prior to her retention and that her conduct demonstrated a degree of settled purpose to remain here.

This weighs in favor of a finding that Maria was habitually resident in the United States on the retention date. The shared parental intent that Maria would choose her country of residence during the summer of 2003 further bolsters this conclusion. Consequently, we hold that Maria is a habitual resident of the United States and that she was not wrongfully retained under Article 3 of the Hague Convention.

For the foregoing reasons, we will affirm the District Court.