Ray Thornton, Justice.
The question presented by this appeal concerns the right of a noncustodial parent, in cases involving disputes over a child’s surname, to insist that the child bear his surname. Appellant Kara Kathleen Huffman, mother of Jacob Austen Huffman (Fisher), brings this appeal of the decision of the Cross County Chancery Court to change her son’s surname to that of his father. We reverse the decision of the chancellor and remand for further findings consistent with this opinion.
Appellant Kara Kathleen Huffman was sixteen years old and unmarried when she gave birth to a son on May 18, 1996. She named her son Jacob Austen Huffman and filed a Certificate of Birth with the Arkansas Department of Health listing appellee John Nicholas Fisher (“Nick”) as the father. In August, 1996, the Arkansas Office of Child Support Enforcement filed suit on behalf of Kara against Nick for child support. Nick filed a third-party complaint in which he admitted that he was the child’s father, and he asked that child support be set and reasonable visitation be established. He also requested that the child’s surname be changed to Fisher.
At a hearing on April 23, 1997, before the Cross County Chancery Court, several witnesses testified concerning whether the child’s surname should be changed from his custodial parent’s surname to his father’s surname. The trial court summarized the testimony and its findings in a letter opinion filed on June 3, 1997. The trial court found that Nick had not paid any child support since Jacob’s birth except for $100.00, although his parents had paid a portion of Kara’s lying-in expenses. The trial court also found that Nick had encouraged Kara, a Catholic, to have an abortion, and that he had counseled her to keep her condition from her parents. Furthermore, there was testimony that Nick had become angry with Kara for getting pregnant and that he had ridiculed her physical appearance during the pregnancy. Since Jacob’s birth, the Fishers had exercised visitation with Jacob in their home on alternate Saturdays from 10:00 a.m. until 6:00 p.m.
Nick testified at trial that he wanted Jacob’s surname to be changed to Fisher because “that’s how I grew up” and he didn’t think he, as a father, should be treated any differently because he had a child out of wedlock. He wanted the child’s name changed because there was the possibility that Kara would marry in the future and take her husband’s name, leaving Jacob with a name different from his mother’s. Finally, he stated that Jacob would be better labeled with a different name from the Huffman family that would be raising him.
Nick’s uncle testified that the Fisher family was a good family and that it would be the proper thing for Jacob to bear the Fisher name, although he could think of no advantage or disadvantage to having one name as opposed to the other. Nick’s father testified that it would be awkward for Nick to explain to others why the child bore his mother’s last name, but that such a scenario probably wouldn’t arise for Kara, and that it was only right for the child to be known as Fisher.
Kara Huffman testified that Nick had not provided her with any financial or emotional support during her pregnancy and that she had made the decision to name Jacob with her family name because he would be raised in her family and would spend his life with her. Kara further offered to retain her maiden name in the event she chose to marry in the future if it would be in Jacob’s best interest.
The trial court noted in its letter opinion that it was initially inclined to change the child’s name to Fisher based upon concerns that Jacob would experience stigma in adolescence The court also noted the applicability of our decision in Reaves v. Herman, 309 Ark. 370, 830 S.W.2d 860 (1992), where we held that there must be compelling facts to show that it would be in the best interest of the child to change his surname. Although the trial court acknowledged that under Reaves there appeared to be no compelling reason to change Jacob’s name to Fisher, it was “still convinced that, in spite of the above, that it would be in Jacob’s best interest that his surname be Fisher.”
The trial court concluded with the following explanation:
When Jacob gets older, he will be faced with the task of explaining to his friends why he does not have his father’s name. If Kara never marries and has no other children, perhaps this would not be so difficult nor embarrassing. However, if Kara remarries [sic], she may very well follow a time-honored tradition and take her husband’s name, although she is certainly not required to do so. If Jacob retains the Huffman surname, he will then not have the name of his biological father nor his biological mother. If Kara remarries [sic] and does not change her surname and she has another child, will she give that child the surname of Huffman so that he/she will have the same name as Jacob or will she give that child the surname of her husband? Certainly most, if not all, of the above is speculation. However, I believe it should be given weight. I still believe that it would be less confusing and embarrassing for Jacob if he took his father’s name . . .
A child needs a surname he can connect with for a lifetime; taking the surname of the mother opens up too many opportunities for the child to be later left without that connection.
* * * *
[T]he Reaves case is a very strong case in favor of Kara . . . We should concentrate on the use of which name will cause Jacob less embarrassment and require him to do less explaining to his friends during his adolescence and young adulthood . . .
It is my opinion that there is a compelling reason to change Jacob’s surname to Fisher. That compelling reason is stated above . . . Although this [children with the mother’s surname] happens often, it is not the norm in this locale. The norm in this locale is that the child will have the same surname as the father. Whether that is right or wrong, that is the norm in this locale. Without the surname of Fisher, Jacob will be faced with the task of explaining to his peers why his name is different. If Jacob keeps the surname of Fluffinan, and if Kara marries and takes the name of her husband, Jacob no longer has the surname of either parent. What is of importance to this court is that Jacob have a surname that he is most likely to be able to connect with one of his parents. The most likelihood of that happening is for Jacob to have the surname of Fisher.
Kara Huffman, a minor, and Jacob Huffman, by and through her parents and the guardians of the person, brought this appeal to the Arkansas Court of Appeals, alleging that the trial court erred in finding that the surname of Jacob should be changed from Huffman to Fisher because the father had failed to present any compelling facts to show that it would be in Jacob’s best interest to change the surname he had carried since birth, and requesting that the best interest rationale adopted by this court in Reaves and in McCullough v. Henderson, 304 Ark. 689, 804 S.W.2d 368 (1991) be clarified by the court to adopt a presumption in favor of the surname chosen for a child by the child’s custodial parent.1 The court of appeals affirmed the trial court by a tie vote, and this court accepted review of the case pursuant to Ark. Sup. Ct. R. 1-2(e)(i). We conduct our review as though the case had been originally appealed to this court. Ark. Sup. Ct. R. 1-2(e).
The use of surnames dates from the Norman conquest, when the growth of population and the development of cities required a means of distinguishing between individuals with identical given names. The Normans also introduced a number of social practices, such as the imposition of a feudal land system and the use of primogeniture as a system of inheritance, that likewise spurred the development of surnames. Gubernat v. Deremer, 140 N.J. 120, 126-127, 657 A.2d 856, 859-60 (1995)(citing Beverly S. Seng, Like Father, Like Child: The Rights of Parents in their Children’s Surnames, 70 Va. L. Rev. 1303, 1323 (1984)). The surname, however, was not as a rule inherited from the father, but was adopted or bestowed based upon one’s occupation, place of habitation, appearance, or other characteristics. Sometimes surnames expressing kinship were chosen, but “patronymics, a name derived from that of the father, was neither compelled nor universal.” Gubernat, 140 N.J. at 128, 657 A.2d at 860 (citing Smith v. U.S. Cas. Co., 197 N.Y. 420, 90 N.E. 947 (1910)). Indeed, it was not uncommon for male children, either illegitimate or the son of a highly respected or dominant married or widowed woman, to be spoken of as belonging to the mother. Id., (citing Seng, supra, at 1321-22).
It was only after the fourteenth century that surnames began to serve as hereditary family names, in part due to the contingency of inheritance of property upon an heir’s retention of the surname associated with that property. Gubernat, 140 N.J. at 129, 657 A.2d at 861 (citing Richard H. Thornton, Note, The Controversy Over Children’s Surnames: Familial Autonomy, Equal Protection and the Child’s Best Interests, 1979 Utah L. Rev. 303, 305 (1979)). A married woman in those times could not maintain suit in her own name; the male was the legal representative of the family. “The custom of patrilineal succession seems to have been a response to England’s medieval social and legal system, which came to vest all rights of ownership and management of marital property in the husband.” Id., (citing In re Schiffman, 28 Cal. 3d 640, 620 P.2d 579, 169 Cal. Rptr. 918 (1980)).
“Allowing the husband to determine the surname of their offspring was part of that system, wherein he was the sole legal representative of the marriage, its property, and its children.” Gubernat, 140 N.J. at 129, 657 A.2d at 861. “Given the secondary status afforded to women at those times, it is not surprising that the masculine lineage was chosen.” Id., (citing M.D. v. A.S.L., 275 N.J. 530, 646 A.2d 543 (1994)). “Inevitably, the institutionalized tradition of assuming the hereditary patronymic surname, and the secondary legal status of women in England, diminished the importance of the maternal surname.” Id., 140 N.J. at 130, 657 A.2d at 862. Review of history demonstrates that the American colonies’ continuation of the English patronymic custom was intrinsically linked to the historical assumption of an inferior social status of women, who belonged legally to their fathers first, then to their husbands. Married women were legally considered the equivalent of children: they could not sue or be sued, deal in property, or be bound on a contract. Gubernat, 140 N.J. at 132, 657 A.2d at 862.
However, a distinction in English common law arose with respect of a child born of unmarried parents. At common law, an illegitimate child wasfilius nullius, the son of no one, or fillius populi, the son of the people. The child had no mother or father recognized by law, and therefore had no legal rights. Because the child could not inherit property, the impetus to bear the paternal surname was diminished. Custom did not dictate the name by which an illegitimate child would be known; the child bore the name gained by reputation in the community.
Id., (citing D.R.S. v. R.S.H., 412 N.E. 1257 (Ind. Ct. App. 1980)).
By the nineteenth century, reform had begun in the consideration of the status of illegitimate children, by replacing the “child of nobody” with a legal family unit of mother and child by way of legislation awarding custody of the child to the mother, consistent with her duty to support him, as his natural guardian. Gubernat, 140 N.J. at 133, 657 A.2d at 864 (citing Secretary of Commonwealth v. City Clerk of Lowell, 373 Mass. 178, 366 N.E.2d 717 (1977)). These statutes made as law that which was already the custom — that a child born of unmarried parents would assume the mother’s surname. Id. “The broader effect of the nineteenth-century statutes was to create divergent treatment of children based upon their birth status. Children born of wedded parents received the paternal surname; children born of unwed parents received the maternal surname.” Id., 140 N.J. at 135, 657 A.2d at 864. “This assumption of matriarchal surnames paralleled the then traditional view that an unmarried woman possessed greater rights to the child as opposed to the putative father.” Id., (citing M.D. v. A.S.L., 275 N.J. at 533, 646 A.2d 543).
Fortunately, society and law have made great strides in the twentieth century, and women have overcome many of the antediluvian attitudes associated with the traditional forms of legal subordination. The General Assembly has eliminated many gender-based differences in marital and parental rights, and refocused our consideration in accordance with each child’s best interests. See, e.g., Ark. Code Ann. § 9-13-101 (Repl. 1998). The best interest of the child is determinative of most all considerations given to a court’s intrusion into the life of a child and his parents, and whl remain the dispositive consideration of this court on the issue at hand, whether the child’s surname should be changed from Huffman, the name he has borne since birth, to Fisher, at the insistence of the father. The best interest rationale has been applied by this court in two recent cases involving the surnames to be given to a child born out of wedlock.
In McCullough v. Henderson, 304 Ark. 689, 804 S.W.2d 368 (1991), Joani McCullough and Mitchell Henderson had a son out of wedlock. Joani named the child Cody McCullough and Mitchell petitioned the court to change the child’s last name to Henderson. The chancellor considered Ark. Code Ann. § 20-18-401(e)(3) (Supp. 1989), which read: “In any case in which paternity of a child is determined by a court of competent jurisdiction, the name of the father and surname of the child shall be entered on the certificate of birth in accordance with the finding and order of the court.” Ark. Code Ann. § 20-18-401(e)(3). Pursuant to this statute, the chancellor concluded that he had no choice but to give the baby its father’s name. McCullough v. Henderson, 304 Ark. 689, 690, 804 S.W.2d 368 (1991).
On appeal, we explained that:
While we agree with the chancellor that the word “shall” renders the provision mandatory, we do not read the statute as directing that the surname of the child should necessarily become that of the father. We think the statute merely states that the full name of the father and the surname of the child shall be entered on the birth certificate “in accordance with the finding and order of the court.” Nothing in the language suggests the two must be the same. In some cases the father may not even want the two names to agree. While we can conceive of instances where the child should bear the father’s name, we can conceive of as many others wherein the welfare of the child, particularly one of more advanced years, would not be well served. We believe a rule which makes the result automatic would be neither prudent nor consistent with the established traditions of the law, hence, we are unwilling to adopt a construction of the statute which produces rigidity, where such an interpretation is decidedly less than self-evident.
Id., 304 Ark. at 691, 804 S.W.2d at 369. Because the chancellor misinterpreted the relevant statute, we remanded the case for a hearing to determine the factors favoring the use of either surname. Id.
A year later, we spoke to the issue again in Reaves v. Herman, 309 Ark. 370, 830 S.W.2d 860 (1992), where we affirmed a lower court finding that there were not compelling facts shown to justify changing a child’s surname from that chosen by his mother, his custodial parent, to that of his father, who sought the name change as well as a determination of paternity. In Reaves, Carrie Herman and Chad Reaves, both seventeen, had a baby boy out of wedlock. Carrie named her son Mitchell Herman. Chad later petitioned the court to change the child’s surname to Reaves, and the chancellor denied his petition. Reaves, 309 Ark. at 372, 830 S.W.2d at 860.
On appeal, we reiterated that Ark. Code Ann. § 20-18-401(e)(3) did not require the child to bear the father’s surname, and that it was a discretionary decision based on the child’s best interest. We affirmed the chancellor’s decision to allow the child to keep his mother’s surname because the father was unwilling to accept the child until sometime after his birth; the mother testified that she had no immediate plans to marry; the mother was willing to retain her maiden name after marriage if that was a concern; and the child had borne the Herman name for nine months prior to the original order. Id. The opinion concluded
Based on the evidence presented at trial we can hardly agree with the appellant that the court abused its discretion since there are no compelling facts that show it would be in the best interest of the minor child to change the surname he has carried since birth. Accordingly, we affirm.
Id., 309 Ark. at 373, 830 S.W.2d at 861.
Most recently, the Arkansas Court of Appeals faced a similar question in Mathews v. Oglesby, 59 Ark. App. 127, 952 S.W. 2d 684 (1997), concerning a mother’s appeal of a chancery court decision to change her son’s surname from that chosen by her at his birth, Mathews, to Oglesby, that of his father. The chancellor in that case was reversed for relying upon a single consideration: the age of the child, to the exclusion of other considerations. Mathews v. Oglesby, 59 Ark. App. 127, 130, 952 S.W.2d 684, 685 (1997). The court of appeals wrote: “The mechanical application of such a policy precludes consideration of the full panoply of factors inherent in determining the best interest of a child.” Rather, “when the best interests of a child are at stake, the chancellor should look into the peculiar circumstances of each case and act as the welfare of the child appears to require.” Id.
Although we adopted the best interest rationale in Reaves and in McCullough, we have not provided sufficient guidelines to trial courts on “the full panoply of factors inherent in determining the best interest of a child,” so as to promote uniformity in the application of the law. The body of case law surrounding this inquiry has produced a compelling list of factors to be considered when addressing a petition to change the surname of a minor child. We hold that in determining the child’s best interest, which must ultimately be the controlling consideration in any change in status, the trial court should consider at least the following factors: (1) the child’s preference; (2) the effect of the change of the child’s surname on the preservation and development of the child’s relationship with each parent; (3) the length of time the child has borne a given name; (4) the degree of community respect associated with the present and proposed surnames; (5) the difficulties, harassment, or embarrassment that the child may experience from bearing the present or proposed surname; and (6) the existence of any parental misconduct or neglect. Barabas v. Rogers, 868 S.W.2d 283 (Tenn. Ct. App. 1993); Daves v. Daves, 105 Wash.2d 24, 31, 711 P.2d 314, 318 (1985); In re Saxton, 309 N.W.2d. 298 (Minn. 1981); In re Marriage of Schiffman, 28 Cal. 3d 640, 620 P.2d 579, 160 Cal. Rptr. 918 (1980); In re Harris, 160 W. Va. 422, 236 S.E.2d 426 (1977)(Harschbarger, J., concurring). See also 57 Am. Jur. 2d Name §§ 46-54 (1988).
We further hold that in order to successfully petition to change a minor child’s surname, the moving party has the burden of demonstrating that such a change is in the best interest of the child. While appellants have not specifically requested that we overrule Reaves, in response to their request to clarify the best interest rationale we have established factors for consideration by the trial court, and we have adopted the clearly erroneous standard of review. These holdings necessarily modify and clarify our holding in Reaves. Where a full inquiry is made by the chancellor of the implication of these factors and a determination is made with due regard to the best interest of the child, the chancellor’s decision will be upheld so long as it is not clearly erroneous. A finding is clearly erroneous when, although there is evidence to support it, the reviewing court on the entire evidence is left with a definite and firm conviction that a mistake has been committed. RAD-Razorback Ltd. Partnership v. B.G. Coney Co., 289 Ark. 550, 713 S.W.2d 462 (1986)(citations omitted).
In his letter opinion, the chancellor addressed the fifth factor at length: the difficulties, harassment, or embarrassment that the child may experience from bearing the present or proposed surname. He then found that the child would suffer less embarrassment and humiliation in adolescence if he took his father’s surname. The trial court’s finding, however, was based almost exclusively upon the chancellor’s perception of the “norm in the locale”; specifically, that Jacob would suffer embarrassment if he were known as Huffman because this was not the prevailing custom in the community. The record, however, does not reflect that any evidence was presented with regard to the “norm in the locale,” and in the absence of testimony presented or judicial notice of the fact that it was indeed such a custom in the area for illegitimate children to take their father’s surname, we can only surmise that the chancellor must have based his conclusion on his own opinion or observations. It is well-settled that a judge cannot be both a witness and a judge in one proceeding. See Ark. R. Evid. 605 (“The judge presiding at the trial may not testify in that trial as a witness. No objection need be made in order to preserve the point.”).
In any event, the “norm in the locale” is not one of the factors to be considered in determining the child’s best interest, although such evidence may be relevant in determining whether the child may experience difficulties, harassment, or embarrassment from bearing a particular surname. See e.g., Clinton v. Morrow., 220 Ark. 377, 247 S.W.2d 1015 (1952)(affirming chancellor who found it was in the best interest of children to have their surname changed from their biological father’s name to their mother’s new married surname to spare the children confusion and embarrassment at school because their name was different from that of their mother). It is important here to re-emphasize that it is always the interest of the child that is paramount. The argument that it might be awkward for Nick to explain why Jacob hears the name Huffman does not carry any weight in supporting a name change. We also note that the trial court reflected on how Jacob might be affected if Kara were to marry and take her husband’s surname or have other children by her husband who might bear a different name from Jacob. Certainly, marriage and having other children in the future is a possibility for both Kara and Nick. However, such reflection on possible future events and their potential impact on Jacob is mere speculation.
In a second point on appeal, appellant asks us to adopt a presumption in favor of the surname chosen for a child by the child’s custodial parent. We have previously held that “a rule which makes the result automatic would be neither prudent nor consistent with the established traditions of the law.” McCullough, 304 Ark. at 691, 804 S.W.2d at 369. Likewise, we decline to do so in this case, on the grounds that such an inflexible resolution will not serve the best interests of the children involved. An individualized determination by the chancellor of the specific facts and relationships through thoughtful and careful consideration of the factors outlined above is required to resolve the question.
While the record before us on appeal might be sufficient to support a reversal and dismissal based upon a de novo review of the record in light of the six factors we have established, we are unwilling to do so without allowing the chancellor an opportunity to address the issue of changing Jacob’s name from Huffinan to Fisher following a full consideration of the holdings in this case. Accordingly, we reverse and remand for further consideration consistent with this opinion.
Reversed and remanded.
Arnold, C.J., Glaze and Corbin, JJ., dissenting.