Sometime around 1962, researchers at Bell Labs—Victor Vyssotsky, Douglas McIlroy, and Robert Morris, Sr.—came up with a computer game they called Darwin. In this game, the players had to write computer programs that fought for domination of a designated memory region. As described in a magazine article in 1972, the object of the game was survival; the programs ("organisms") had the ability to "kill" each other, and could create copies of themselves. This article is the earliest published resource that I have witnessed to use the term virus in the context of self-replicating software. Specifically, the text mentions that one of the players "invented a virus—an unkillable organism" that was able to win several games due to the way it protected itself from attacks launched by adversary programs.

The virus reference in the game of Darwin doesn't quite match our understanding of what a traditional virus is; however, it does provide a perspective on the origins of early self-replicating programs. By the way, if you are a trivia buff, you might be interested to know that the cocreator of Darwin, Robert Morris, Sr., is the father of Robert Tappan Morris, Jr., who is the author of the infamous Internet Worm. Keep that one handy the next time you play Trivial Pursuit!

An article published in 1984 by A. K. Dewdney popularized a version of Darwin under the name Core War. In Dewdney's game, computer programs "stalk each other from address to address.… Sometimes they go scouting for the enemy; sometimes they lay down a barrage of numeric bombs; sometimes they copy themselves out of danger or stop to repair damage." Like modern viruses, programs in Core War and Darwin were designed with replication in mind, although they did not have the parasitic properties that we have come to associate with typical virus specimens today.

The first confirmed implementation of self-replicating code that existed in the wild as part of a host program was PERVADE, written by John Walker in 1975. PERVADE was a general-purpose routine that could be called by any program that required propagation capabilities. According to Walker, when PERVADE was invoked, "It created an independent process which, while the host program was going about its business, would examine all the directories accessible to its caller. If a directory did not contain a copy of the program, or contained an older version, PERVADE would copy the version being executed into that directory". I guess that's why they called it PERVADE; it permeates the system using this technique.

The only program known to host PERVADE was ANIMAL—Walker's implementation of a popular game in which the computer tries to guess which animal the player has in mind. Walker's version of the game was significantly better than many other versions, and people kept asking him for copies. Looking for an innovative way to distribute the software, he coupled ANIMAL with the PERVADE routine. The resulting program possessed viral properties that allowed it to spread from directory to directory. Furthermore, when users exchanged tapes containing "infected" copies of the game, it propagated to other systems. Although people didn't use the word virus at that time to describe such software, there was a connection to the term nonetheless: The program's source code included a variable named VIRUS to control whether the PERVADE routine should be activated.

The early 1980s presented the world with a series of viral programs built for Apple II personal computers. The most notorious of these is Elk Cloner, written in 1982 by high school junior Rich Skrenta. Skrenta recalls that he enjoyed "playing jokes on schoolmates by altering copies of pirated games to self-destruct after a number of plays". According to him, Elk Cloner was an attempt to impact the friends' disks without having physical access to them. To achieve this goal, he crafted the program to reside in a floppy disk's boot sector, and become active when the system booted up from the infected disk. Elk Cloner would then load into memory, and copy itself to new disks whenever they were inserted into the computer. Every once in a while, the program would display the following lyrical message:

ELK CLONER:

 THE PROGRAM WITH A PERSONALITY

IT WILL GET ON ALL YOUR DISKS

IT WILL INFILTRATE YOUR CHIPS

YES IT'S CLONER!

IT WILL STICK TO YOU LIKE GLUE

IT WILL MODIFY RAM TOO

SEND IN THE CLONER!

It's quite clear that young Skrenta was more of a software developer than a poet. However, at least he could rhyme, and the meter isn't half bad. Beyond such linguistic nit-pickings, though, his pathogenic code was quite successful, spreading far and wide by the standards of its time.

Another viral program for Apple II was created independently around the same time by Joe Dellinger, a student at Texas A&M University. This was mainly a proof-of-concept program that resided in the boot sector and kept track of the number of floppy disks it had infected. Like Frankenstein's monster, Dellinger's creation did not receive an official name, and people now refer to several of its versions simply as Virus 1, Virus 2, and Virus 3.

The security community did not commonly start using the word virus to refer to such programs until 1984, when Fred Cohen offered his definition of the term to the public in a research paper titled "Computer Viruses—Theory and Experiments." Cohen's pioneering work formally examined the phenomenon of self-replicating software, described the significance of the threat associated with viruses, and pointed out that "Little work has been done in the area of keeping information entering an area from causing damage". Some sources credit his seminar advisor, Len Adleman, with assigning the term virus to Cohen's concept. (Yes, that's Len Adleman who is the "A" in RSA, the famous public key cryptographic algorithm. What a small world!)

It is generally accepted that the first virus that targeted Microsoft DOS computers was discovered in the wild in 1986. It was called the Brain virus, mainly because it changed the label of infected diskettes to say "(c) Brain." Like the Apple II viral programs before it, Brain spread by attaching itself to the floppy disk's boot sector. An early version of Brain included the following "advertisement," which led researchers to believe that the virus was authored by Basit and Amjad Farooq Alvi :

Welcome to the Dungeon

(c) 1986 Basit & Amjad (pvt) Ltd.

BRAIN COMPUTER SERVICES

730 NIZAB BLOCK ALLAMA IQBAL TOWN

LAHORE-PAKISTAN

PHONE :430791,443248,280530.

Beware of this VIRUS....

Contact us for vaccination............ $#@%$@!!

Virdem was another Microsoft DOS virus that appeared in 1986, and was developed independently of Brain. It was written by Ralf Burger as a demonstration program for the Chaos Computer Club conference to help explain the functionality of a computer virus. Unlike its predecessors, which relied on the disk's boot sector to propagate, Virdem spread by attaching to files that had the .COM file extension.

The programs that we have covered in this brief historical overview are summarized belowGiven the lack of definitive records that document the dawn of viruses, keep in mind that this is not an exhaustive list of early viral software. Consider this a sampling of influential specimens with origins that can be traced with a moderate degree of certainty.