Davis v. State

476 N.E.2d 127 (1985)

Reed DAVIS and Mary Davis, Appellants (Defendants below), v. STATE of Indiana, Appellee (Plaintiff below).

No. 2-1083A356.

Court of Appeals of Indiana, Second District.

March 28, 1985.

*130 J.J. Paul, III, James H. Voyles, Indianapolis, for appellants.

Linley E. Pearson, Atty. Gen., Lee Cloyd, Deputy Atty. Gen., Indianapolis, for appellee.

SHIELDS, Judge.

Defendants Reed Davis and Mary Davis (Davises) appeal their convictions for neglect of a dependent, Ind. Code Ann. § 35-46-1-4 (Burns 1979),[1] a class D felony. They raise the following issues:

1) whether Ind. Code § 35-46-1-4 is unconstitutionally vague; 2) whether the indictments were sufficient; 3) whether the trial court properly admitted evidence of parentage probabilities; and 4) whether the evidence was sufficient to support the convictions.

Judgment affirmed as to Mary and reversed as to Reed.

On August 3, 1982, at approximately 12:30 p.m., a full-term male infant (Baby Lucky),[2] estimated to be a few hours old, was found at the side of a gravel road near Warren, Indiana. An anonymous telephone call to the police implicated the Davises as the parents. After an investigation and blood tests to determine the probabilities of parentage, the Davises were indicted by the Huntington County grand jury for neglect of a dependent. After a jury trial, both were found guilty and sentenced to an executed term of two years.

I. CONSTITUTIONALITY

The Davises argue Ind. Code § 35-46-1-4(a)(1) is unconstitutionally vague. Specifically, they allege the prohibition of conduct which "places the dependent in a situation that may endanger his life or health," Ind. Code § 35-46-1-4(a) fails to inform the public and law enforcement officers of the specific conduct prohibited. Ind. Code § 35-46-1-4(a) (emphasis added). In their brief, the Davises focus on the use of the permissive word "may". For example, they argue parental permission to engage in interscholastic and contact sports "may endanger" a child's life or health. Consequently, they allege the statute proscribes even reasonable conduct where life or health is not actually endangered.[3]

Thus, the Davises attempt to present a facial challenge to the constitutionality of the neglect statute. However, "it is well established that vagueness challenges to statutes which do not involve First Amendment freedoms must be examined in light of the facts of the case at hand." United States v. Mazurie, 419 U.S. 544, 550, 95 S. Ct. 710, 714, 42 L. Ed. 2d 706 (1975). The Davises are not at liberty to devise a hypothetical situation which would demonstrate vagueness;[4] the statute is *131 void for vagueness only if it is vague as applied to the precise circumstances of their case. Id.; accord United States v. Powell, 423 U.S. 87, 96 S. Ct. 316, 46 L. Ed. 2d 228 (1975); e.g., Terrel v. State, 170 Ind. App. 422, 353 N.E.2d 553 (1976) (statute omitting "knowledge" from offense of visiting a common nuisance not vague where evidence demonstrated defendant's knowledge); Connell v. City of Logansport, 397 N.E.2d 1058 (Ind. App. 1979) ("conduct unbecoming an officer" not vague where defendant addressed his superior in obscene, abusive and threatening language). In making this determination we are not concerned with the ability of our Legislature to have chosen clearer and more precise language. United States v. Powell, 423 U.S. at 94, 96 S. Ct. at 320. Rather, we must determine whether an individual of ordinary intelligence would reasonably understand that his contemplated conduct is proscribed. United States v. Mazurie, 419 U.S. at 553, 95 S. Ct. at 715.

Although the Davises argue the statute proscribes even reasonable conduct which may not actually endanger the dependent, such conduct was not involved in the case at bar. The evidence indicates that within a few hours of birth and with the umbilical cord still attached, Baby Lucky was wrapped in a brown paper sack and left by the side of a deserted road in a wooded area outside the view of passersby. A jogger found the baby in a dehydrated condition, sunburned and bruised with superficial cuts and scratches and a puncture wound on the back of his head. No reasonable person of ordinary intelligence would have difficulty determining that such activity was proscribed by the statute at issue. Given the nature of the Davises conduct, the statute was sufficient to advise the Davises that their treatment of Baby Lucky was proscribed.

II. INDICTMENTS

The Davises were individually indicted for neglect of a dependent. The indictments were identical except for the named defendant:

"On or after the 3rd day of August, 1982, in Huntington County, State of Indiana, [Reed/Mary] Davis, then having the care, custody or control of an infant male child, a dependent, did knowingly place such dependent in a situation that did endanger his health and/or did abandon said dependent child."

Record at 14, 16. The indictments cited Ind. Code § 35-46-1-4 as the statutory provision allegedly violated. The Davises allege 1) the indictments fail to state any facts comprising the basis of the offense charged and 2) a violation of due process in the charging of both endangerment and abandonment in the same indictment.[5]

The Davises unsuccessfully sought dismissal of the indictments in the trial court because "the indictment does not set out that situation and circumstances which allegedly endangered the life or health of Baby Lucky." In support of their argument, at trial and on appeal, the Davises cite Ind. Code § 35-34-1-2(a) (Burns Supp. 1984) (formerly 35-3.1-1-2(a)) which requires the indictment to contain "the essential facts constituting the offense charged." However, the Davises waived the sufficiency of the indictments by failing to include the issue in their motion to correct errors.

Nonetheless, we find the indictments were sufficient. Omitting the formal parts, both indictments specified the time and place of the alleged offense, identified the defendants by name, and identified them as yet unnamed victim as "an infant male child, a dependent," all in language paralleling the statute proscribing neglect of a dependent. The indictments then followed the statutory language in specifying the particular acts which constitute *132 the alleged neglect. When the statutory language enumerates the specific acts which constitute the crime, an indictment paralleling the words of the statute is sufficient. Allison v. State, 240 Ind. 556, 166 N.E.2d 171 (1960), cert. denied, 365 U.S. 608, 81 S. Ct. 822, 5 L. Ed. 2d 821 (1961). There is no need for further description of the injury or methods employed in committing the crime. Moody v. State, 448 N.E.2d 660 (Ind. 1983).

Nor did the charging of a violation of both subsections (a)(1) and (a)(2) render the information duplicitous. We first observe the charge of endangerment "and/or" abandonment in the indictment effectively charged the Davises with both acts; the use of "or" was superfluous. In State v. Pratt, 255 La. 919, 233 So. 2d 883 (1970), the Supreme Court of Louisiana found a similar indictment sufficient. Pratt challenged a charge of aggravated rape where the victim's resistance was alleged to have been overcome by force "and/or" threats of great bodily harm; the aggravated rape statute listed force and intimidation in two disjunctive subsections, either of which constituted aggravated rape. Under the conjunctive rule, the prosecution was allowed to charge disjunctive acts in conjunctive language because the prosecution may have been unavoidably uncertain which form of "and/or" was equivalent to charging Pratt with both acts, in the conjunctive only, and did not prevent Pratt from understanding the charge. We similarly conclude the Davises were charged with both endangerment and abandonment; the inclusion of "and/or" did not hamper their ability to understand the nature of the charge or to prepare a defense.

Indiana also adheres to the conjunctive rule which allows the prosecution to charge disjunctive acts in conjunctive language. The neglect of a dependent statute is framed in disjunctive subsections. The crime of neglect of a dependent occurs by the commission of any one of the five distinct acts, including 1) endangerment, 2) abandonment, 3) confinement, 4) deprivation of support, or 5) deprivation of education.[6] Ind. Code § 35-46-1-4. Here, the indictments enumerated two acts constituting neglect in conjunctive language, endangerment and abandonment. Accordingly, the indictments did not create a hybrid offense, nor were they duplicitous:

"[W]here the statute denounces several acts as a crime, they may be charged in one indictment or in a single count if they are connected in the conjunctive. An indictment drawn in that manner is not duplicitous, and it suffices to prove any one or more of the charges."

Shanholt v. State, 448 N.E.2d 308, 314 (Ind. App. 1983) quoting United States v. Amick, 439 F.2d 351 (7th Cir.), cert. denied 403 U.S. 918, 91 S. Ct. 2227, 29 L. Ed. 2d 694 (1971). The grand jury indicted the Davises for neglect of a dependent based upon two acts, either of which constitutes neglect. Under such an indictment, the State needed only to prove the Davises either endangered or abandoned Baby Lucky.

III. PARENTAGE

The Davises also allege the trial court erred in overruling their motion in limine and continuing objections which sought to exclude expert testimony concerning the mathematical probability of the Davises' parentage of Baby Lucky based on the results of blood tests. Specifically, the Davises argue 1) the potentially exaggerated impact of probability evidence, 2) the invalidity of "Bayes Theorem" to calculate the probabilities involved, 3) the misrepresentation of calculations as the probability of parentage rather than the probability of exclusion, and 4) the absence of any foundation for probability calculations when neither parent is known.

The target of the Davises objections was the expert testimony of Michael Conneally, a professor of medical genetics at Indiana University. Conneally based his testimony on the results of two separate series of standard tests: 1) HLA (Human Leucocyte *133 Antigen) tissue-typing and 2) a series of blood tests for 20 separate genetic marker systems. The test results are not disputed; Only the statistical presentation of the probability calculations based upon the results are the subject of argument.

Conneally first testified that the results of the tests "do not exclude the possibility that Mr. and Miss [sic] Davis are the father and mother are the parents of Infant Doe." Record at 409. Conneally then testified regarding the "probability of parentage" computed by multiplying the frequency of occurrence of each genetic marker in the caucasian population of Indiana. The frequencies were based on accepted scientific tables. Record at 417-18. Based on the 20 series test,[7] Conneally concluded "the probability that uh Mr. and Miss [sic] Davis are the parents of Baby Lucky rather than another couple picked at random from the community is ninety-eight point three percent (98.3%)." Based on the Davises rare HLA typing, Conneally testified:

"the chance that they are the parents of this child versus a random set of parents in the community ... comes out to be ninety nine point nine eight eight six seven percent (99.98867%) or just based on tissue typing alone odds of eight thousand eight hundred and twenty eight (8,828) to one that Mr. and Mrs. Davis are the parents of Baby Lucky rather than another couple taken from the community... ."

Record at 423. Conneally then combined the results of the HLA and 20 series test and concluded:

"we have ninety-eight point seven percent (98.7%). .. . When we multiply these we get odds of five hundred and eighteen thousand (518,000) to one (1). .. . [T]he probability of ninety-nine point three nine eight 0 [sic] seven percent (99.398079%) that Mr. and Mrs. Davis are the father [sic] of Baby Lucky rather than another couple taken at random in the community."

Record at 423-24. Conneally converted the probability percentages into "real numbers" and concluded only one couple in half a million couples could have been the parents of Baby Lucky.

The Davises also presented expert testimony regarding the mathematical probabilities of parentage. Dr. Herbert Polesky, Director of the Minneapolis War Memorial Blood Bank tested for 15 different genetic marker systems; again, none of the test results excluded either Reed or Mary as the parents of Baby Lucky.

Polesky testified "there was a 91.4% greater chance that Reed and mary could have been the parents" rather than a random couple. Record at 553. On cross-examination, Polesky opined, using the results of both his 15 series test and Conneally's 20 series test, the Davises "would be ninety-nine percent (99%) more likely to have been the parents" than a random couple, which in turn represented that 1 out of every hundred couples could also be the parents. Record at 566.

Polesky then converted the results into "real numbers". Based on his 15 series test, one out of every ten couples could have parented Baby Lucky. If Conneally's 20 series results were added, one out of every 91 random couples could be the parents. Finally, upon adding the HLA results, 1 out of every 10,000 couples could have produced Baby Lucky. Discounting for age, the local population contained approximately 125,000 to 130,000 potential parent-couples.[8] Polesky then explained that compared to another non-excluded couple,[9] based on the blood tests alone, the *134 chances that the Davises were the parents was 50/50.

Probability Testimony

The Davises challenge the probability evidence as unduly prejudicial. They claim the jury was necessarily but "unlawfully impressed by the mystique of mathematical demonstrations." Consequently, they argue the probability evidence should have been excluded because of its "potentially exaggerated impact." Appellant's Brief at 37.

Courts and commentators have traditionally viewed mathematical probability testimony with extreme caution because of its need for foundational support and its need for sufficient explanation to the fact finder.

Foundational deficiencies are widespread. For example, in the leading case of People v. Collins, 68 Cal. 2d 319, 438 P.2d 33, 66 Cal. Rptr. 497 (1968) the California Supreme Court reversed a conviction for robbery based on the unduly prejudicial effect of mathematical probabilities introduced by the prosecutor. A mathematics instructor at a state college had calculated the probability that the couple charged committed the robbery based on the occurrence in the population of the couple's unusual physical characteristics.[10] The prosecutor in closing argument thereby arrived at a probability of one in 12 million that any randomly selected couple possessed the charged couples' combination of distinctive characteristics. However, the probabilities assigned to each individual characteristic were purely speculative. As emphasized by the court: "Without presenting any statistical evidence whatsoever in support of the probabilities for the factors selected, the prosecutor then proceeded to have the witnesses assume probability factors for the various characteristics which he deemed to be shared by the guilty couple and all other couples answering to such distinctive characteristics." Id. at 325, 438 P.2d at 36-37, 66 Cal. Rptr. at 500-501, (emphasis in original). Moreover, even the testimony on which the calculations were based, describing the physical characteristics of the robbers, was conflicting. The court concluded that although "risks of error permeate the prosecution's circumstantial case, ... few jurors could resist the temptation to accord disproportionate weight" to a "numerical index of probable guilt." Id. at 330, 438 P.2d at 40, 66 Cal. Rptr. at 504.

Similarly inadequate statistical bases for probability calculations have been consistently rejected in criminal cases. In United States v. Massey, 594 F.2d 676 (8th Cir.1979) an expert witness testified to the similarity of hair samples taken from the defendant and found in a ski mask allegedly used in a bank robbery. The expert testified the hair samples matched, and he had been unable to distinguish the hair of two individuals only twice in 2,000 cases. During the expert's testimony, the trial judge attempted to convert the testimony into a statistical probability of one in a thousand. The witness denied the validity of converting his personal experience into statistical probabilities. Nonetheless, the prosecutor, in closing argument, exacerbated the trial judge's misunderstanding of the evidence and argued that a "better than 99.44%" chance was proof beyond a reasonable doubt. The prosecutor's misuse of the already confusing colloquy was held to be prejudicial error. Accord, Miller v. State, 240 Ark. 340, 399 S.W.2d 268 (1966) (estimated probabilities of occurrence of soil types); see generally Annot., 36 A.L.R.3d 1194 (1971). Thus, the concern of these courts which resulted in ultimately excluding probability testimony focuses not on the calculations but upon the foundation for the calculations. When unsubstantiated estimates are used in probability calculations, speculation is presented to the jury clothed in scientific accuracy; the prejudicial impact clearly outweighs the probative value.

*135 However, where probability testimony is based on empirical scientific data, rather than unsubstantiated estimates, the presentation and admission of probability testimony need not constitute error. Three recent cases emphasize this distinction between an inadequate and an adequate foundation for probability testimony. In State v. Washington, 229 Kan. 47, 622 P.2d 986 (1981), an expert witness was permitted to testify in a murder prosecution that only .6% of the population had blood with the same characteristics as the defendant. Similarly, in State v. Garrison, 120 Ariz. 255, 585 P.2d 563 (1978), a dentist was permitted to testify there was an "eight in one million probability" that teeth marks found on the victim's body were not made by the defendant. And in People v. DiGiacomo, 71 Ill. App.3d 56, 27 Ill.Dec. 232, 388 N.E.2d 1281 (1979), a criminologist was allowed to testify that the chances a hair sample belonged to a person other than the defendant was "1 in 4500." In all three cases, a sufficient foundation demonstrated the reliability of the probability calculations based upon scientific reports or tables rather than upon mere speculation.

There is no foundational error in the instant case. The probability calculations were not based upon speculation but upon accepted scientific tables reporting the frequency of each genetic marker in the caucasian population in Indiana. Both expert witnesses employed the same method of calculation and arrived at virtually identical probability percentages.[11]

We are also unpersuaded by the Davises claim of "exaggerated impact" due to the alleged ability of seemingly impressive numbers to mislead or confuse the jury. Nor are we persuaded probability calculations are too difficult or complicated to explain.

The Davises brief and our independent research reveal only one case which found error in the admission of probability calculations which were unquestioningly based upon empirical scientific data. In State v. Carlson, 267 N.W.2d 170 (Minn. 1978) an expert witness for the State testified there was a "1 in 800 chance" and a "1 in 4,500 chance" the hair samples did not belong to the accused. The Supreme Court of Minnesota expressed concern over the potentially exaggerated impact on the trier of fact: "Testimony expressing opinions or conclusions in terms of statistical probabilities can make the uncertain seem all but proven, and suggest, by quantification, satisfaction of the requirement that guilt be established `beyond a reasonable doubt'." Id. at 176, citing Tribe, Trial by Mathematics, 84 Harv.L.Rev. 1329 (1971). Although the court acknowledged the ability of diligent cross-examination to minimize confusion, the psychological impact of mathematical precision was found too compelling. The court nonetheless found the testimony cumulative of another expert witness's testimony that the samples were similar and concluded the erroneous admission of probability calculations was not prejudicial. In State v. Boyd, 331 N.W.2d 480 (Minn. 1983), using the same rationale, the supreme court of Minnesota excluded statistical probability evidence in paternity actions.

The approach taken in Minnesota, however, has been rejected by an impressive myriad of courts and commentators. Indeed, blood test results have enjoyed increasing acceptance in the judicial forum because of the increased reliability of HLA testing procedures to exclude a significant portion of the population as possible fathers in paternity actions.

In 1976, the American Bar Association and the American Medical Association jointly promulgated guidelines governing the use of blood and tissue tests in paternity actions. See Abbott, Sell & Krause, Joint AMA-ABA Guidelines: Present Status of Serologic Testing in Problems of Disputed Parentage, 10 Fam.L.Q. 247 (1976). The guidelines specifically recommended *136 admission of evidence relating to the probability of paternity:

"[We recommend] that the National Conference of Commissioners on Uniform State Laws develop new legislation or amend the `Uniform Parentage Act' and the `Uniform Blood Test Act' to ... simplify the admissibility in evidence of test results and the probative effect thereof, including the evidentiary value of estimations of `likelihood of paternity'."

Id. at 283. Indeed, the Uniform Parentage Act now expressly provides for the admission of the statistical probability of paternity generated by Bayes Theorem, the formula used in the instant case:

"Evidence relating to paternity may include: ..... (3) blood test results, weighted in accordance with evidence, if available, of the statistical probability of the alleged father's paternity."

Uniform Parentage Act § 12. The Commissioners' Comments to the Uniform Parentage Act explain that when a parent is not excluded, a "paternity index" is calculated which in turn enables the experts to calculate the probability of paternity. If the probability exceeds 95% or falls below 5%, the results have produced "de facto inclusions or exclusions." Id. at 604.

The Indiana Legislature has recognized the significant medical advances in inclusionary testing procedures and has liberalized the use of such evidence in paternity actions. See Garfield, 1980 Survey of Recent Developments in Indiana Law, 14 Ind.L.Rev. 358 (1981). In 1980, Ind. Code Ann. § 31-6-6.1-8 (Burns 1980), was amended to allow the admission of test results into evidence.[12] Indeed, a majority of states now admit the results of blood and tissue typing tests, not only to exclude the alleged father, but also as evidence on the issue of paternity where the alleged father is not excluded. E. Cleary, McCormick on Evidence 621-22, 657 (3d ed. 1984).

In most jurisdictions, the expert testimony is not limited to a finding of similarity, or "nonexclusion". Id. at 657. The significance of the similarity of blood types cannot be intelligently understood without knowledge of the frequency of that blood type in the relevant population. "Where reasonable estimates of the population frequencies are available, they should not be kept from the jury." Id. at 655. Typically, in paternity cases, experts report the frequency of distribution of the incriminating genetic markers in the relevant population. Id. at 657. This testimony is often characterized as the "probability of exclusion," in other words, the proportion of the relevant population that the tests would exclude *137 based upon the frequency of occurrence of the incriminating characteristics. Id.; see, Hausner v. Blackman, 7 Kan. App. 2d 693, 648 P.2d 249 (1982) aff'd, 233 Kan. 223, 662 P.2d 1183 (1983).

Thus, testimony regarding the probability of exclusion, without more, has been viewed as incomplete and misleading. E. Cleary, supra at 657-58. In Hausner v. Blackman, 648 P.2d at 253-54, for example, the court reversed a jury finding of paternity where an expert testified to a 70% probability of exclusion without calculating the probability of paternity. In other words, the series of tests run by the expert in Hausner were designed to exclude 70% of falsely accused males; however, although the defendant was not excluded the expert failed to calculate the probability of his paternity based upon the individual results derived in the defendant's case. Accordingly, although Kansas excludes testimony regarding the probability of exclusion, in Tice v. Richardson, 7 Kan. App. 2d 509, 644 P.2d 490 (1982) an expert was permitted to testify to a 99.96% "plausibility" of paternity.

Several other jurisdictions have allowed testimony regarding the probability of paternity. In Carlyon v. Weeks, 387 So. 2d 465 (Fla. Dist. Ct. App. 1980), the expert was allowed to testify to a "plausibility of paternity" of 99.9%. In California and Missouri, experts were allowed to testify to probabilities of paternity of 85.95% and 81.15%, respectively, although these percentages were considered inconclusive. County of Ventura v. Marcus, 139 Cal. App. 3d 612, 189 Cal. Rptr. 8 (1983); State v. Williams, 609 S.W.2d 456 (Mo. Ct. App. 1980).

We recognize the need for caution in the admission and evaluation of probability testimony. The presentation of a percentage representing the probability of parentage may be deceptively attractive to the fact finder. Indeed, it may be well within the trial court's discretion to caution the fact finder against equating the probability of parentage with proven guilt. However, thorough preparation and cross-examination, as demonstrated in the instant case, are adequate opportunities for clarification. For example, if the probability of parentage is 99%, a full 1% of the relevant population, or 1 out of every 100 persons tested at random, would also exhibit the incriminating characteristics. In a relevant population of 1,000,000 persons, 10,000 persons would also have a probability of parentage of 99%. Again, based on the blood tests alone, the defendant is no more likely than any of the remaining 10,000 persons to be the true parent.[13] In short, the defendant is entitled to explain he is merely one member of the relatively small population which exhibits the incriminating genetic markers. When explained in this manner, the fact finder need only further understand that the probability of parentage does not consider the nonquantitive evidence which the jury must evaluate together with the probability evidence.

We agree with the Supreme Court of Utah in its evaluation of the jury's ability to evaluate this probability testimony: "This Court ... [has] a higher opinion of a jury's ability to weigh the credibility of such figures when properly presented and challenged, and accord this type of testimony the weight it deserves." State v. Clayton, 646 P.2d 723, at n. 1 (Utah 1982). A jury which finds this presentation of this type of evidence in conjunction with the nonquantitive evidence to be highly persuasive is justified in its evaluation.

Bayes Theorem

Both experts used Bayes Theorem, to which the Davises object, in assigning percentages to the Davises' probability of parentage. The use of Bayes Theorem is mathematically necessary to convert the probabilities into percentages. Record at 558. In their calculations, both experts assumed an equal chance the Davises or another random couple were the parents. *138 The Davises contend this assumption improperly ignored the weight of the other circumstantial non-test evidence and denied them the presumption of innocence.

The function of Bayes Theorem, a probability theory, is to show the effect of a new item of evidence on a previously established probability. In this case the new item was the blood and tissue tests. The previously established probability was the probability of their parentage based on the other non-test evidence without the benefit of the scientific tests. Theoretically, Bayes Theorem permits a mathematical calculation of the probability of the Davises' parentage of Baby Lucky based upon all the information known about Baby Lucky and the Davises.

Simplified, the theorem reduces to the multiplication of the probability of the Davises' parentage based on the circumstantial non-test evidence ("prior probability") times the likelihood of the Davises' parentage based on the scientific tests ("likelihood ratio"). The product of these two factors represents the total probability of the Davises parentage of Baby Lucky based on all the evidence available. This simplification may be expressed as follows: (PROBABILITY BASED UPON NON-TEST EVIDENCE) X (PROBABILITY BASED UPON SCIENTIFIC TESTS ONLY) = TOTAL PROBABILITY OF PARENTAGE. The product is expressed in percentile form. In the instant case, the prosecution's and the defendant's experts arrived at a total probability of parentage of 99.98867% and 99.% respectively.

In arriving at their respective percentages, both experts were restricted to a consideration of the test evidence only. In order to employ the Theorem, the experts therefore substituted a neutral probability for the prior probability variable, i.e., the probability based upon non-test evidence. The precise probability employed was "50/50", which reduces to a factor of 1, and was described as an equal chance that the Davises or a couple chosen at random parented Baby Lucky.[14] The neutrality of this substitution can be demonstrated as follows: (1) x (99.98867) = 99.98867%. In other words, the application of Bayes Theorem to the scientific tests, with a prior probability of 1 or 50/50, determined the probability based exclusively upon the tests. The 50/50 assumption was completely neutral.

The Davises' argument therefore defeats itself. They contend if they had been incarcerated at the time of Baby Lucky's birth and abandonment, the calculations employed by the experts would have ignored this fact and therefore the probabilities generated by Bayes Theorem would still have been in the high nineties due to the inclusion of the 50/50 assumption. However, the prior probability generated by these hypothetical facts and theoretically employed by the fact finder would have approached zero. A substitution of this probability by the fact finder would have yielded a total probability of parentage of 0%: (O) X (PROBABILITY BASED UPON BLOOD TESTS ONLY) = 0% PROBABILITY OF PARENTAGE; or (0) X (99.98867) = 0%. In other words, the blood test evidence would have had no impact on the fact finder.

We reject the Davises' invitation to require the experts to include a prior probability based upon a consideration of the circumstantial non-test evidence available concerning the Davises' parentage of Baby Lucky. It is the function of the fact finder to determine the prior probability of the Davises' parentage based upon the non-test evidence introduced at trial. Any expert determination of this prior probability would invade the function of the jury. It properly remained a function of the fact finder to determine the Davises' prior probability and, ultimately, the total probability of parentage.

*139 Probability of Exclusion

Similarly, both experts represented the calculations as the probability of parentage rather than the probability of exclusion. The Davises cite to no testimony or authority which impeaches the characterization of the results as the "probability of parentage."[15] Further, the evidence was that all couples who shared Baby Lucky's blood and tissue characteristics would have an equal probability of parentage.

Foundation

Lastly, we reject the Davises challenge to the scientific community's acceptance of probability calculations where neither parent is known. Although Conneally testified this was the first occasion on which he calculated such probabilities in a criminal case, he had made such calculations "many times in non-criminal cases for research." Record at 433. He specifically denied such calculations where neither parent was known was an experimental protocol. As the Davises own expert testified, the same method of calculation was typically used by colleagues in kidnapping and immigration cases. Although neither expert had been previously called upon to compute parentage probabilities where neither parent was known, the method of calculation was demonstrated to be generally accepted by the scientific community. See, e.g., Kuo, Linking a Bloodstain to a Missing Person by Genetic Inheritance, J. Forensic Sciences, 438 (1982). The trial court did not err in admitting the probability evidence.

IV. SUFFICIENCY

Both Reed and Mary allege the evidence was insufficient to sustain their convictions. We find the evidence sufficient to sustain Mary's conviction but insufficient to convict Reed.

Mary

Mary alleges the evidence was insufficient to demonstrate she was the mother of Baby Lucky and therefore she was not shown to have "the care, custody or control" of Baby Lucky. See Ind. Code Ann. § 35-46-1-4(a). In reviewing sufficiency questions we do not reweigh the evidence or judge the credibility of witnesses. Rather we consider only the evidence most favorable to the State to determine whether there was substantial evidence of probative value to support the conviction. Smith v. State, 408 N.E.2d 614 (Ind. App. 1980).

In the case at bar, the evidence most favorable to the State reveals Baby Lucky was born on August 3rd, 1982 and was a few hours old when he was found at approximately 12:30 p.m. on that date. Mary underwent a physical examination on August 11, 1982, eight days after Baby Lucky was born. The examining witness physician testified Mary exhibited an enlarged uterus and pendulous breasts with a milky discharge among other indications of a recent pregnancy. Although Mary claimed she had experienced "flooding" or a miscarriage on the night of August 2d, 1982, the examining physician concluded Mary's condition was consistent with the recent delivery of a "close to full-term" baby one to three weeks earlier. Although Mary testified she took birth control pills, she filled only one prescription on September 11, 1981, 11 months before Baby Lucky was born.

Several witnesses testified Mary suffered from swollen legs and ankles and appeared to be pregnant in the days and months immediately preceding Baby Lucky's birth. Co-workers testified Mary appeared very pale but much slimmer when she appeared for work late in the afternoon on August 3rd.

Blood tests indicated a 99.98867% probability Reed and Mary were the parents of *140 Baby Lucky. The H.L.A. results revealed both Baby Lucky and Tyson, the Davises' young son, had identical H.L.A. typing. Expert testimony indicated that essentially no two people exhibit identical H.L.A. types unless they are closely related such as brother and sister. We find the evidence sufficient to support the finding that Mary was Baby Lucky's mother.

Mary also alleges the evidence was insufficient to demonstrate she abandoned or placed Baby Lucky in a situation that endangered his health. Baby Lucky was born just a "few hours" before he was found between 12:30 p.m. and 1:00 p.m. on August 3rd. The uncontroverted evidence indicates Mary was home alone, accompanied only by her son Tyson, between the hours of 5:30 a.m. and 3:30 p.m. on the day Baby Lucky was born. Sometime shortly after birth, Baby Lucky was left alone by the side of a deserted country gravel road out of the view of passersby. The evidence of Mary's sole control over Baby Lucky supports a reasonable inference that she abandoned him or placed him in a situation that endangered his health. We must therefore affirm Mary's conviction.

Reed

The evidence is similarly sufficient to conclude Reed was Baby Lucky's father. However, Ind. Code § 35-46-1-4 requires the endangerment or abandonment to have been either knowing or intentional. Although a parent need not possess a specific intent to commit neglect, he or she must at a minimum be "aware of facts that would alert a reasonable parent under the circumstances to take affirmative action to protect the child." Smith v. State, 408 N.E.2d at 621.

Even assuming Reed was aware of Mary's pregnancy, the uncontradicted evidence placed Reed at his workplace when Baby Lucky was born, abandoned, and found.[16] Reed's workplace was a twenty minute drive from his home. He clocked in at work at 6:08 a.m. and clocked out at 5:40 p.m. He took his lunch to work and was observed at his workplace throughout the day. We must therefore reverse Reed's conviction for insufficient evidence.

Judgment affirmed with respect to Mary and reversed with respect to Reed.

BUCHANAN, C.J., concurs.

SULLIVAN, J., concurs, with separate opinion.

SULLIVAN, Judge, concurring.

I fully concur in the majority opinion as to Issues II through IV. However, I disagree with the conclusion under Issue I that the Davises are prohibited from challenging the overbreadth of I.C. XX-XX-X-X(a). The statute is overbroad as to any person prosecuted under it. It is, therefore, overbroad as to the Davises. To say that the Davises may not assert the overbreadth because their alleged conduct actually endangered the child is to beg the question. The fact remains that the statute does not prohibit conduct which actually endangers, unless an interpretation more restrictive than the language would suggest is taken.

Furthermore, the statute is so vague as to provide law officers with no guidelines or standards for its enforcement. In my view, this would ordinarily constitute a fatal defect. Kolender v. Lawson (1983) 461 U.S. 352, 103 S. Ct. 1855, 75 L. Ed. 2d 903; People v. Berck (1973) 32 N.Y.2d 567, 347 N.Y.S.2d 33, 300 N.E.2d 411.

The defect, however, is not fatal in this instance. As indicated by Kolender v. Lawson, supra, state courts may place a limiting construction upon an otherwise overbroad statute and thus preserve its constitutionality. It is therefore within our prerogative and I would exercise it to construe the word "may" in I.C. XX-XX-X-X(a) to mean "is likely to." Corn Products Refining Co. v. F.T.C. (1945) 324 U.S. 726, 65 S. Ct. 961, 89 L. Ed. 1320. As so construed, *141 the statute gains the requisite definiteness and properly subjects the defendants here to prosecution for the acts charged.

Subject to this qualification, I concur.

NOTES

[1] Ind. Code 35-46-1-4(a) provides as follows:

"A person having the care, custody, or control of a dependent who knowingly or intentionally: (1) places the dependent in a situation that may endanger his life or health; (2) abandons or cruelly confines the dependent; (3) deprives the dependent of necessary support; or (4) deprives the dependent of education as required by law; commits neglect of a dependent. .. ."

[2] The infant was alternatively referred to as "Infant Doe" and "Baby Lucky" at trial.

[3] We note "endanger" is defined as "danger or peril of probable harm or loss." Webster's Third New International Dictionary (1976) (emphasis added).

[4] The facial invalidity of a statute for vagueness or overbreadth may be raised outside the facts of the case in controversy where free speech and other first amendment rights are implicated. See, e.g., Kolender v. Lawson, 461 U.S. 352, n. 8, 103 S. Ct. 1855, n. 8, 75 L. Ed. 2d 903 (1983); Connell, 397 N.E.2d at 1064, n. 2.

[5] Ind. Code § 35-46-1-4(a) lists five separate forms of neglect, two of which are endangerment and abandonment. See supra note 1.

[6] Items 2) and 3) are contained in a single statutory subsection.

[7] The blood tests identified different blood characteristics, or "genetic markers." Conneally tested for 20 different characteristics.

[8] Upon the Davises motion, the trial court took judicial notice of the area population at 438,719.

[9] The use of a "random man" is an accepted practice. As defined in E. Cleary, McCormick on Evidence (3d ed. 1984), the random man is "a hypothetical entity whose genotypes (and hence phenotypes) are a kind of average across all [excluded and non-excluded] men. Id. at 660. Indeed, "the use of the random man factor is inherent in the HLA paternity test." Tice v. Richardson, 7 Kan. App. 2d 509, 644 P.2d 490, 494 (1982).

[10] The crime was allegedly committed by a caucasian woman with a blond ponytail who left the scene accompanied by a black male with a beard and a mustache in a partly yellow automobile.

[11] The two experts differed significantly only in the conversion of the probability percentages into "real numbers;" The Davises not only did not object to such a conversion, but routinely employed the conversion to their advantage.

[12] In 1980, Ind. Code Ann. § 31-6-6.1-8 read:

"Upon motion of any party, the court shall order all of the parties to the action to undergo either a blood grouping test or a Human Leukocyte Antigen (HLA) tissue test. The tests shall be performed by a qualified expert approved by the court, and the results of the tests may be received in evidence." (emphasis added)

In 1983, the statute was amended, in part, to admit both the test results and the expert's finding.

"(a) Upon the motion of any party, the court shall order all of the parties to the action to undergo blood antigen testing. The tests shall be performed by a qualified expert approved by the court. The results of the tests, together with the finding of the expert constitute conclusive evidence if the results and finding exclude a party as the biological father of the child. The results and finding are admissible in all paternity proceedings, unless the court excludes the results or findings for good cause." (emphasis added.)

The current version, as amended in 1984, redesignated the subsections as follows:

"(a) Upon the motion of any party, the court shall order all of the parties to the action to undergo blood testing. The tests shall be performed by a qualified expert approved by the court.

(b) The results of the tests, together with the finding of the expert constitute conclusive evidence if the results and finding exclude a party as the biological father of the child. The results and finding are admissible in all paternity proceedings, unless the court excludes the results or finding for good cause.

As observed by at least one commentator, "the statutory amendments reflect continuing confidence in the validity of blood testing." King, 1983 Survey of Recent Developments in Indiana Law, 17 Ind.L.Rev. 173, 195 (1984).

[13] Consequently, the similarity of blood types or the probability of paternity is as insufficient to support a finding of paternity as is mere presence at the scene of a crime insufficient to support a conviction for theft.

[14] Moreover, as stated by the Davises' own expert witness, the initial assumption of equal probabilities of parentage between any given couple and the remaining population is a standard assumption "always" used in probability calculations. E. Cleary, supra note 8 at 659, n. 16.

[15] We recognize the probabilities were capable of alternative phrasing. For example, a 99% probability of parentage also indicates 1% of the population, or 1 out of 100 couples, could have parented Baby Lucky. This alternative view was properly and adequately explained to the jury through the expert testimony, as indicated, and affected the weight, rather than the admissibility, of the calculations.

[16] We express no opinion on whether Reed's conduct after Baby Lucky was found constituted Nonsupport of a Dependent Child under Ind. Code Ann. § 35-46-1-5. Support is defined in Ind. Code Ann. § 35-46-1-1 (Burns Supp. 1984) as "food, clothing, shelter or medical care."