concurring in part and dissenting in part.
I concur in the majority’s decision that claims 37-40, 45, and 69 of the '444 patent do not require automatic computer determination of teeth finish positions, and that the law of the case doctrine bars Align’s cross-appeal. I respectfully dissent, however, from those portions of the majority’s decision that affirm the district court’s grants of summary judgment to Align on the issues of non-infringement and non-enablement. I believe that the district court failed to construe properly the relevant claim terms, and improperly analyzed and shifted the burden of proof on the issue of enablement. I further believe that an insufficient record exists to support an attempt to correct the district court’s omissions by construing the claims on appeal. I find further that the majority’s decision, while acknowledging the need to construe the claims, improperly imports limitations from both the specifications' and an ancestor patent into many of those claims. For these reasons, I would reverse and remand.
I.
Align’s principal argument, implicitly accepted by the majority, is that neither the district court, nor this court on review, need begin the infringement analysis with the language of the claims. Instead, Align argues that an examination of the specification of the patents in suit and the prosecution history of an antecedent patent permits a court to find that the heart of Ormco’s invention is the practice of automatically determining finish tooth positions and to compare the accused product to that finding, rather than to the claims of the governing patents. I disagree.
*1321At the outset, it is apparent that the trial court improperly considered the question of infringement without actually construing the claim terms at issue. Indeed, the district court specifically stated that it, “was not interpreting the specific language of the claims to favor one side or another.” Ormco Corp. v. Align Tech., Inc., No. 03-cv-00016, slip op. n. 1 (C.D.Cal. May 13, 2004).1 Though this court has undertaken claim construction in the first instance, it has also remanded cases for further construction in the absence of any meaningful claim construction by the district court. The court has rejected the assertion that, because it conducts a de novo review of claim construction, a district court’s claim construction “does not matter” and the absence of a claim construction is not error. Nazomi Commc’n, Inc. v. Arm Holdings, PLC, 403 F.3d 1364, 1371 (Fed.Cir.2005). In Nazomi, the court explained, in remanding for further claim construction, that “[t]his court’s review of a district court’s claim construction, albeit without deference, nonetheless is not an independent analysis in the first instance. Moreover, in order to perform such a review, this court must be furnished ‘sufficient findings and reasoning to permit meaningful appellate scrutiny.’ ” Id. (citing Gechter v. Davidson, 116 F.3d 1454, 1458 (Fed.Cir.1997)). The district court here has furnished this court with no such findings. See also Shuffle Master, Inc. v. Vending-data Corp., 163 Fed.Appx. 864, 868-69 (Fed.Cir.2005) (“While claim construction is a question of law, the district court’s analysis is important to the process of claim construction, and in this context, as in others, we decline to construe the claim without the guidance of the district court’s construction.”); Anchor Wall Sys. v. Rockwood Retaining Walls, Inc., 340 F.3d 1298, 1311 (Fed.Cir.2003) (finding: “[i]n order to review the court’s finding of non-infringement, we must know what meaning and scope the district court gave to the asserted claims”); Graco, Inc. v. Binks Mfg. Co., 60 F.3d 785, 791 (Fed.Cir.1995) (stating that the “entire omission of a claim construction analysis from the opinion ... provide[s] an independent basis for remand”).
While I sympathize with the district court’s desire to resolve the case before it without resort to the daunting task of construing'ninety-two claims in four separate patents, a non-infringement decision which eschews the very exercise of claim construction is inconsistent with this court’s repeated directives to district courts.2 I believe, accordingly, that the district court’s failure to construe the claim language in this case, standing alone, warrants remand.
II.
Rather than remand the matter, the majority chooses, instead, to conduct its own *1322infringement analysis. It opens that analysis by recognizing that claim construction begins with the language of the claims themselves. In support, the majority cites Phillips and Markman for the truisms that, “the claims of a patent define the invention to which the patentee is entitled the right to exclude,” and that, though claims, “must be read in view of the specification, of which they are a part,” it is the language of the claims that is to be construed, not the specification. Phillips v. AWH Corp., 415 F.3d 1303, 1312 (Fed.Cir.2005) (en banc); Markman v. Westview Instruments, Inc., 52 F.3d 967, 979 (Fed.Cir.1995) (en banc), aff'd, 517 U.S. 370, 116 S.Ct. 1384, 134 L.Ed.2d 577 (1996). The majority, nevertheless, appears to ignore its own opening admonitions in the remainder of its opinion. First, the majority does not “begin” with the language of the claims. Indeed, it never actually, tells us what word or words in the claims in suit it purports to construe. At best, after examining the specification of one of the patents in suit (but not its prosecution history), and the prosecution history of a patent which is not in suit, it backs into a form of claim construction by asserting that all of the claims address themselves to the practice of determining finish tooth positions. Nowhere, however, does the majority tell us what language is used in which claims to describe that practice, or why such language is in need of interpretation.
The majority concedes, as it must, that none of the claims in suit “expressly recite automatic control of the finish tooth positioning.” Despite the absence of this language in the claims, however, the majority concludes that this unstated (and seemingly important) limitation is “what [the claims] mean.”
In its analysis of the specification, the majority first cites to general statements in the Background and Summary of Invention sections of the patent where the inventors discuss the benefits of computer-assisted (ie., automated) methods for designing orthodontic appliances, most particularly the greater accuracy that can be gained by reducing reliance on human decision-making. It then examines the preferred embodiments and concludes that those embodiments only contemplate human or operator input into the process at the front end^ — ie., when the user first inputs the data into the computer for processing. From these portions of the specification, the majority concludes that the essence of Ormco’s invention requires, “automatic determination of the finish positions of teeth without human adjustment of the final results.” This exercise of attempting to glean the intended invention from sources other than the claims themselves is precisely what Align asked that we do. (See Appellee’s Br., 36, 39 (“Ormco made it very clear in the specification ... that the very nature of its invention was for automatic computer determination of the finish positions of teeth ...;” and “the statements Align drew from the specification and the file histories are admissions concerning the fundamental nature of Ormco’s invention, not narrow statements concerning the meaning of some particular specific term.”) (emphases in original).)
This court, however, has rejected a claim construction process based on the “essence” of an invention. See, e.g., Allen Eng’g Corp. v. Bartell Indus., Inc., 299 F.3d 1336, 1345 (Fed.Cir.2002) (“It is well settled that ‘there is no legally recognizable or protected essential element, gist or heart of the invention in a combination patent.’ ”) (citing Aro Mfg. Co. v. Convertible Top Replacement Co., 365 U.S. 336, 345, 81 S.Ct. 599, 5 L.Ed.2d 592 (1961)). Indeed, this court has done so quite forcefully and quite recently. See MBO Labs., Inc. v. Becton, Dickinson & Co., 474 F.3d 1323, 1330-31 (Fed.Cir.2007) (“We sympathize with the district court’s choice, since *1323we agree that [the feature] is an essential element of the invention.... However, we cannot endorse a construction analysis that does not identify ‘a textual reference in the actual language of the claim with which to associate a proffered claim construction.’ ”) (quoting Johnson Worldwide Assocs., Inc. v. Zebco Corp., 175 F.3d 985, 990 (Fed.Cir.1999)). Simply stated, “automatic determination of finish tooth positions without human adjustment of the final results” is a limitation that the majority has amalgamated from the specification of one of the patents in suit without reference to the specific language of any claim of any of the patents.
The court’s recent decision in Ventana Medical Systems, Inc. v. Biogenex Laboratories, Inc., 473 F.3d 1173 (Fed.Cir.2006), counsels against this practice. In Ventana, the issue was the proper construction of the term “dispensing” in a patent claiming automated methods for staining microscope slides. Ventana, 473 F.3d at 1176. The district court construed “dispensing” to require “direct dispensing,” because the embodiments in the specification involved direct dispensing (much as the district court here determined that processing, determining, and calculating orthodontic solutions require automatic processing, automatic determining, and automatic calculating). Id. at 1178. On appeal, Biogenex argued that the specification, when read in its entirety, would lead to the “inescapable conclusion” that the heart of the invention involved “direct dispensing,” and that the specification implicitly defined the term “dispensing” to mean “direct dispensing.” This court, in Venta-na, rejected that narrowed construction of the claims, noting this court’s previous repeated warnings against confining claims solely to disclosed embodiments. Id. at 1181 (“[While] the fact that the disclosed, embodiments are limited can assist in interpreting claim language ... [it] does not in and of itself mean that the method claims at issue are limited to the disclosed embodiments.”); see also Phillips, 415 F.3d at 1323 (rejecting, “the contention that if a patent describes only a single embodiment, the claims of the patent must be construed as being limited to that embodiment”). The court should do so again here.
As mentioned above, there are four separate patents in suit, the '444, the '861, the '243, and the '432. Ormco points to ninety-two claims in those patents which it claims Align has infringed, seventy-six of which are found in the '444 patent. By way of example, independent claim 1 of the '444 patent claims:
A method for use in the orthodontic correction of malocclused teeth of a patient in accordance with the individual anatomy of the patient and a prescription of an orthodontic practitioner for treatment of the patient, the method comprising:
generating data by scanning shapes of the teeth of a patient;
displaying a graphic representation of the teeth of the patient with a computer from the generated data; and
through an operator interacting with a computer located remote from the orthodontic practitioner, altering the graphic representation to arrange a plurality of the teeth in relation to each other in accordance with the prescription, to produce a digital model of a desired arrangement of the teeth of the patient that includes data of the shapes of a plurality of the teeth positioned relative to each other.
'444 patent col.67 1.57-col.68 1.6. Though this claim does not exclude operator involvement in the process and, indeed, claims the involvement’ of an operator in certain steps of the process, and though the claim never uses the term “automatically,” the majority concludes that this *1324claim, and virtually every other claim in suit, claims a process that is “completely automatic” and allows for no skilled operator involvement. In short, the majority has: imported the terms “automatically” and “automated” from the specification to the claims; found that the terms “automatically” and “automated,” when used in the specification, are the same terms (without a record of how one skilled in the art would construe those terms);3 concluded that, in every instance, the words “automatically” and “automated” mean “completely automatically” or “completely automated;” found that, where a claim is silent on the issue, use of an operator is wholly precluded; and found that, where involvement of an operator is claimed, that operator cannot be skilled. Given the limited record in the trial court, I cannot find support either in the claims or in the record for these determinations.
In reaching these conclusions, moreover, the majority necessarily ignores those places in the claims where the addition of an automatic process is discussed. For instance, claim 4 of the '444 patent recites: “The method of claim 3 further comprising: generating machine code in response to the designed geometry; and transmitting the machine code to a manufacturing machine to substantially automatically operate the machine in accordance with the shapes of the plurality of teeth.” '444 patent, col. 68, 11. 19-26 (emphasis added). Why would a claim use the phrase “substantially automatically operate the machine” if the inventors intended that every claim included a requirement that the machine be operated “completely” or “wholly” automatically? Who or what completes the “substantial” operation of the machine if not an operator? These kinds of questions emphasize the need for an assessment of the actual language used in the claims, rather than a construction of the invention that is reached despite that language.
Apparently recognizing the danger of counseling a district court to rely on language from the specification to the exclusion of language in the claims themselves, and recognizing that the statements in the specifications are less definitive than Align claims, the majority concedes that “standing alone” the statements in the specification may not be conclusive to show that the claims require completely “automatic determination of finish tooth positions.” To support its construction of the invention, accordingly, the majority turns to the prosecution history of the '562 patent and finds that the doctrine of prosecution history disclaimer mandates the conclusion the majority reaches. This approach is similarly unavailing. See Armament Sys. & Procedures, Inc. v. Monadnock Lifetime Prods., 168 F.3d 1319 (Table), 1998 WL 537746, at **3 (Fed.Cir. Aug. 7, 1998) (citing Markman, 52 F.3d at 980) (emphasis added) (noting that “Monadnock next argues that the prosecution history requires that the four-step process be read into claim 1 of the '297 patent. Again we disagree,” and finding that “any argument based on prosecution history must fail for the same reason as the specification argument: lack of textual support in the claim”)-, id. (“Accordingly, statements appearing in the file history of the patent *1325are not sufficient to add entirely new limitations to a claim.”).
Prosecution disclaimer requires a pat-entee to clearly and unmistakably disavow certain interpretations, and I find no such disavowal here. SanDisk Corp. v. Memorex Prods., Inc., 415 F.3d 1278, 1287 (Fed.Cir.2005) (“There is no ‘clear and unmistakable’ disclaimer if a prosecution argument is subject to more than one reasonable interpretation, one of which is consistent with a proffered meaning of the disputed term.”); Omega Eng’g, Inc. v. Raytek Corp., 334 F.3d 1314, 1324 (Fed.Cir.2003). Importantly, while the quotations lifted from the '562 patent history do appear to support the majority’s conclusion that the process contemplated in that potential invention was a highly automated one, that language was proffered to the examiner in connection with claims in that patent which do not share claim language with the majority of the claims at issue in this suit. The amendments were all made in response to the examiner’s concerns regarding the formation of “ideal dental archforms,” and the use of a computer to “derive ... tooth finish positions ... to place the teeth on the ideal dental archform.” That process simply is not claimed in the '444 patent. The majority concedes this fact, but finds no limitations on the application of prosecution disclaimer posed by it. The majority cites Wang Laboratories, Inc. v. America Online, Inc., 197 F.3d 1377, 1384 (Fed.Cir.1999), and Jonsson v. Stanley Works, 903 F.2d 812, 818 (Fed.Cir.1990), for the proposition that statements in the prosecution history of an ancestor patent are relevant as long as the patents in suit address the same general subject matter.
The court’s more recent decisions, however, set forth a more restrictive rule- — • requiring common language or a linguistic “hook” among the claims before resort to a parent application’s prosecution history is appropriate. Invitrogen Corp. v. Clontech Labs., Inc., 429 F.3d 1052, 1078 (Fed.Cir.2005) (“[T]he prosecution of one claim term in a parent application will generally not limit different claim language in a continuation application.”); ResQNet.com, Inc. v. Lansa, Inc., 346 F.3d 1374, 1383 (Fed.Cir.2003) (“Although a parent patent’s prosecution history may inform the claim construction of its descendent ... prosecution history is irrelevant to the meaning of [a] limitation [if] the two patents do not share the same claim language.”); Advanced Cardiovascular Sys., Inc. v. Medtronic, Inc., 265 F.3d 1294, 1305-06 (Fed.Cir.2001) (finding prosecution history of parent patents to be irrelevant where “there are no common claims in dispute”); see also Alloc, Inc. v. Int’l Trade Comm’n, 342 F.3d 1361, 1381 (Fed.Cir.2003) (Schall, J.,- dissenting) (“Statements ... made during the prosecution of a parent application can only apply to continuation applications if the parent and child patents contain the same claim limitations.”); Serrano v. Telular Corp., 111 F.3d 1578, 1584 (Fed.Cir.1997) (stating that “[t]he patentee’s statement concerning whether the prior art discloses a ‘send’ signal means is relevant only to those claims which require the generation of such a signal, and those claims are not asserted here” and reasoning that “[although statements in a file history may of course be used to explain and potentially limit the meaning of claim limitations, ... [they] cannot be used to add an entirely new limitation to the claim”) (emphases added). Here, however, neither Align, the district court, nor the majority has identified any linguistic hooks between the '562 and '444 patents.4
*1326Again, the court’s recent decision in Ventana provides guidance on the issue. In Ventana, this court rejected the argument that, in prosecuting an ancestor patent, the applicants disclaimed certain devices that could not dispense reagent “directly to a sample,” explaining that: “[bjecause claims 1 and 5 of the [descended] patent use different claim language ... the alleged disclaimer [of the ancestor patent] ... does not apply to the asserted claims of the [descended] patent.” Ventana, 473 F.3d at 1182 (emphases added, citations omitted) (also stating that the court examines a patent’s prosecution history to “determine whether the inventor disclaimed a particular interpretation of a claim term during the prosecution of the patent in suit or ... of an ancestor application. But the doctrine of prosecution disclaimer generally does not apply when the claim term in the descendant patent uses different language.”) (emphasis added). In Ventana, the court correctly examined the specific language of the claims, not the general subject matter of the invention, when it looked to the prosecution history of the ancestor patents. Thus, in Ventana, although the claims of the descended patent involved similar subject matter and, indeed, shared some common language, this court nonetheless held that critically different language between the claims precluded the use of disclaiming statements made while prosecuting the ancestor patent to limit the claims of the descended patent. In sum, the only potential relevance of the '562 prosecution in this case would be to illuminate the construction of a word or linguistic hook shared by the patents in suit.5 Invitrogen, 429 F.3d at 1078 stating: “a prosecution disclaimer or prosecution history estoppel argument, ... falters on the principle that the prosecution of one claim term in a parent application will generally not limit different claim language in a continuation application”). Align has not provided those hooks.
III.
The district court’s rulings regarding enablement were based wholly upon this flawed claim construction and, therefore, are unsustainable as well. See Chiron Corp. v. Genentech, Inc., 363 F.3d 1247, 1254 (Fed.Cir.2004) (stating: “an enablement inquiry typically begins with a construction of the claims”). Specifically, the district court found that, because he believed the patent intended a wholly automatic process, and he did not find that the inventors had ever successfully performed the process in the absence of any operator involvement, the invention was not enabled by the patent. Because, as discussed above, the first of the district court’s premises was based on a flawed infringement analysis, its non-enablement conclusion must also be reversed. There are, moreover, additional flaws with the trial court’s analysis of the enablement of Orm-co’s patents that should be noted.
An enablement inquiry turns on whether the specification of a challenged patent: “provide[s] sufficient teaching such that one skilled in the art could make and use the full scope of the invention without undue experimentation.” Warner-Lambert Co. v. Teva Pharm. USA, Inc., 418 F.3d 1326, 1337 (Fed.Cir.2005) (citations omit*1327ted). “Furthermore, ‘[wjhether undue experimentation is needed is not a single, simple factual determination, but rather is a conclusion reached by weighing many factual considerations.’ ” Id. (quoting In re Wands, 858 F.2d 731, 737 (Fed.Cir.1988)). “Some of these considerations, commonly referred to as ‘the Wands factors,’ include ‘(1) the quantity of experimentation necessary, (2) the amount of direction or guidance presented, (3) the presence or absence of working examples, (4) the nature of the invention, (5) the state of the prior art, (6) the relative skill of those in the art, (7) the predictability or unpredictability of the art, and (8) the breadth of the claims.’ ” Id. (quoting Wands, 858 F.2d at 737).
Here, however, the district court did not examine the specifications to determine whether one of ordinary skill in the art could make and use the invention without undue experimentation. Indeed, there is no indication that the district court considered any of the factors enumerated in Wands. Ormco Corp. v. Align Tech., Inc., No. 03-cv-00016, slip op. at 3-7 (C.D.Cal. Aug. 23, 2004). Instead, Align only produced, and the district court only focused on, evidence of whether Ormco had perfected a commercially successful version of the invention. Commercial success, however, is not determinative of enablement. See CFMT, Inc. v. Yieldup Int’l Corp., 349 F.3d 1333, 1338 (Fed.Cir.2003) (“Title 35 does not require that a patent disclosure enable one of ordinary skill in the art to make and use a perfected, commercially viable embodiment absent a claim limitation to that effect.”).6 I believe, therefore, that the district court’s limited, improper examination of the enablement issue could not have supported summary judgment in Align’s favor on this issue.
IV.
For the foregoing reasons, I cannot endorse the district court’s infringement analysis or its analysis of enablement, and cannot agree with the majority’s construction of the claims in the face of an inadequate record upon which to do so. While it ultimately may be true that some language of some of the claims in suit may require the very “construction” the district court and the majority afford them (because the specification is critical to a proper claim construction process, and a parent application does meaningfully inform construction of similar claim language in later patents), it is impossible to tell from the current record whether, and to what extent, that fact is true. It appears, moreover, that, as to many, if not most, of the critical claims at issue, the construction the majority endorses is simply not supported by a proper application of this court’s precedents. Accordingly, I would reverse the district court’s grants of summary judgment, and remand with direction that the claim language at issue be construed and that a record supporting that construction be developed.
Honorable Kathleen M. O’Malley, District Judge, United States District. Court for the Northern District of' Ohio, sitting by designation.
. The district court cited Microsoft Corp. v. Multi-Tech Systems, Inc., 357 F.3d 1340, 1347 (Fed.Cir.2004), as support for its claim construction approach. In Microsoft, however, contrary to the approach taken by the district court here, this court turned to the specification and prosecution history only after an examination of the claim language. See id.
. There are legitimate ways in which district courts can streamline the claim construction analysis when faced with myriad claims from multiple patents. District courts may choose, for instance, to construe only the independent claims of the various patents, or may direct the parties to identify the most representative claims for construction. In that way, the district court can provide guidance as to its construction of the most critical or oft-repeated claim terms and, thus, provide a roadmap with respect to the direction any additional claim construction might take. Employing strategies to streamline the claim construction process is fundamentally different, however, from avoiding the process altogether.
. The absence of a record regarding how one of ordinary skill in the art would have understood the terms "automated” and “automatically” at the time the application was filed is particularly important. In fast-moving technologies, like the computer technology at issue here, words can mean vastly different things at different points in time. Thus, what we would consider to be meaningful advances in the use of computers in a particular process (i.e., injecting automation into a process previously done by hand) today could be far different from what one skilled in the art would understand was possible — and, thus, was novel — even a few years earlier.
. Ormco concedes that there are linguistic hooks to some claims in some of the patents, particularly in the '432. This appears true, however, only as to a minority of the ninety-two asserted claims.
. Ormco also correctly points out that claims 1 and 37 of the '444 patent were allowed in the first office action and that those claims of the '444 patent that were rejected over Lem-chen were allowed after Ormco made amendments having nothing to do with automatically determining the finish positions of teeth. Thus, the majority’s present application of the '562 history, to the exclusion of the '444's own prosecution history, would seem to render the '444's prosecution meaningless.
. It is also apparent that the district court improperly shifted the burden of proof to Ormco by requiring it to produce corroborated testimony on the issue of enablement. "[CJorroboration is required of any witness whose testimony alone is asserted to invalidate a patent, regardless of his or her level of interest.” Finnigan Corp. v. Int’l Trade Comm’n, 180 F.3d 1354, 1369 (Fed.Cir.1999) (emphasis added). Here, Align, not Ormco, sought invalidation of the patent.